CN103729839B - A kind of method and system of sensor-based outdoor camera tracking - Google Patents
A kind of method and system of sensor-based outdoor camera tracking Download PDFInfo
- Publication number
- CN103729839B CN103729839B CN201310596129.9A CN201310596129A CN103729839B CN 103729839 B CN103729839 B CN 103729839B CN 201310596129 A CN201310596129 A CN 201310596129A CN 103729839 B CN103729839 B CN 103729839B
- Authority
- CN
- China
- Prior art keywords
- camera
- parameter
- intrinsic parameter
- cameras
- video camera
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 239000011159 matrix material Substances 0.000 claims abstract description 86
- 230000003287 optical effect Effects 0.000 claims abstract description 59
- 230000036544 posture Effects 0.000 claims description 17
- 238000005259 measurement Methods 0.000 claims description 12
- 230000001133 acceleration Effects 0.000 claims description 8
- 241000208340 Araliaceae Species 0.000 claims description 5
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims description 5
- 235000003140 Panax quinquefolius Nutrition 0.000 claims description 5
- 235000008434 ginseng Nutrition 0.000 claims description 5
- 230000001131 transforming effect Effects 0.000 claims description 4
- 240000007594 Oryza sativa Species 0.000 claims description 2
- 235000007164 Oryza sativa Nutrition 0.000 claims description 2
- 235000009566 rice Nutrition 0.000 claims description 2
- VIKNJXKGJWUCNN-XGXHKTLJSA-N norethisterone Chemical group O=C1CC[C@@H]2[C@H]3CC[C@](C)([C@](CC4)(O)C#C)[C@@H]4[C@@H]3CCC2=C1 VIKNJXKGJWUCNN-XGXHKTLJSA-N 0.000 claims 1
- 230000002708 enhancing effect Effects 0.000 abstract description 6
- 238000012545 processing Methods 0.000 abstract description 6
- 238000010586 diagram Methods 0.000 description 13
- 238000003672 processing method Methods 0.000 description 12
- 230000000694 effects Effects 0.000 description 10
- 238000012937 correction Methods 0.000 description 6
- 230000009182 swimming Effects 0.000 description 6
- 239000000571 coke Substances 0.000 description 5
- 238000005457 optimization Methods 0.000 description 5
- 238000006243 chemical reaction Methods 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 239000011800 void material Substances 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000005304 joining Methods 0.000 description 1
- 239000002932 luster Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000003032 molecular docking Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Abstract
This application provides a kind of methods of sensor-based outdoor camera tracking, comprising: calibrating camera intrinsic parameter, the camera intrinsic parameter include focal length, optical center image coordinate, distortion parameter and holder parameter;External parameters of cameras is obtained, the external parameters of cameras includes orthogonal matrix and optical center world coordinates;Projection matrix is generated according to the camera intrinsic parameter and external parameters of cameras;According to the projection matrix, virtual pattern is broadcasted on caption machine.The present invention improves the precision of camera calibration on the basis of not increasing mark plate area, so that the precision of Camera location is improved, to be suitable for the processing and enhancing of high quality and high efficiency sports relay signal.
Description
Technical field
This application involves the technical field of camera calibration, more particularly to a kind of sensor-based camera calibration with
The method and system of tracking.
Background technique
During television relay, often need on television image be superimposed virtual pattern with help spectators more fully understand and
Appreciate watched TV programme.Especially in the relaying process of competitive sports, for different sports items, need because of ground
The information such as games results, Site characteristic, sportsman's data, sponsor are shown to suiting measures to different conditions on arena, for example, it is desired in track and field
World record line is shown in the match of project, and the national flag being laid on swimming lane is shown in the match of swimming event, and, in foot
Show that offside line and 9 code line of penalty shot etc., these virtual patterns generally require one as actual scene in the match of ball project
Point, it is fused among actual scene.
In order to realize the effect of virtual reality fusion, needs to solve the problems, such as Camera location, that is, need to solve computer real-time
The variation of tracking camera allows void in the hope of in video camera rotation or while zoom, virtual pattern, which also can be carried out, correspondingly to be changed
Quasi- figure and actual scene are bonded always.
In order to solve the problems, such as Camera location, there are mainly two types of method, sensor, method and figures for the technology generallyd use
Image space method.Specifically, sensor, method is to record video camera in real time by the embedded sensors in the gear of camera pan-tilt
Rotation or zoom information, to calculate the corresponding change of virtual pattern;Image method is by the Direct Recognition to scene, meter
Calculate video camera rotation or when zoom scene image space variation, to calculate the corresponding change of virtual pattern.
For sensor, method, need to carry out camera calibration in advance to determine the basic parameter of video camera, camera shooting
The precision of machine calibration will determine the effect of Camera location.When subject is longer to video camera distance, subject
Smaller in picture, in track and field event and swimming event match, the focus value of camera operation is usually 50 meters, works as mark plate
Area reach 10 square metres or more when, the lattice point of mark plate could be told in picture to carry out camera calibration.10 is flat
More than square rice mark plate not only transport difficult, but also make precision and be difficult to ensure, therefore, camera calibration is caused to be difficult to complete
Perhaps not accurate enough situation is demarcated so as to cause Camera location to be difficult to the effect completed or tracked not accurate enough.
For image method, since place is easy to be blocked or be stained by personnel or other objects, easily cause
In practice can not reconnaissance perhaps reconnaissance inaccuracy situation the identification in place is just difficult to complete or identify it is not accurate enough, from
And the superposition of virtual pattern be just difficult to complete or be superimposed it is not accurate enough.
Therefore, a technical problem that needs to be urgently solved by technical personnel in the field at present is exactly: providing a kind of based on biography
The camera calibration of sensor and the method and system of tracking, to improve video camera on the basis of not increasing mark plate area
The precision of calibration, to improve the precision of Camera location, be suitable for the processing of high quality and high efficiency sports relay signal with
Enhancing.
Summary of the invention
The technical problem to be solved by the application is to provide a kind of method of sensor-based outdoor camera tracking and
System, to improve the precision of camera calibration on the basis of not increasing mark plate area, to improve Camera location
Precision, to be suitable for the processing and enhancing of high quality and high efficiency sports relay signal.
To solve the above-mentioned problems, this application discloses a kind of method of sensor-based outdoor camera tracking, packets
It includes:
Calibrating camera intrinsic parameter, the camera intrinsic parameter include focal length, optical center image coordinate, distortion parameter and holder
Parameter;
External parameters of cameras is obtained, the external parameters of cameras includes orthogonal matrix and optical center world coordinates;
Projection matrix is generated according to the camera intrinsic parameter and external parameters of cameras;
According to the projection matrix, virtual pattern is broadcasted on caption machine.
Preferably, described the step of generating projection matrix according to the camera intrinsic parameter and external parameters of cameras it
Before, further includes:
The camera intrinsic parameter and external parameters of cameras are adjusted according to presetting rule, obtain camera shooting adjusted
Machine intrinsic parameter and external parameters of cameras.
Preferably, the step of calibrating camera intrinsic parameter includes:
The initial intrinsic parameter of video camera under fixed camera lens zoom level and focus value is obtained using Zhang Zhengyou algorithm;
Video camera is turned into different postures, and is taken pictures to the mark plate of known dimensions, while recording lower sensor
The data of return;
According to the data that the initial intrinsic parameter of the video camera and sensor return, video camera internal reference is calculated using following formula
Number:
inpara=(fx,fy,u0,v0,kc1,kc2,zAxis,yAxis,φ0)
Wherein, θ, φ are respectively the yawing angle and pitch angle that sensor is passed back;xi,jFor on jth picture, i-th point
Image coordinate, PjFor the corresponding projection matrix of the corresponding video camera posture of jth picture, XiFor i-th point of world coordinates, it is
Known quantity;fx,fyFor the focal length of both direction, u0,v0For the image coordinate of both direction optical center, kc1For distortion 1, kc2For distortion
2, zAxis be z-axis away from, yAxis be y-axis away from, φ0For pitching zero deviation angle.
Preferably, the holder parameter include z-axis away from zAxis, y-axis away from yAxis and pitching zero deviation angle φ0, described
Obtain external parameters of cameras the step of include:
Outer parameter matrix is calculated using following formula according to the holder parameter:
Wherein, orthogonal matrix R33It is obtained by following formula:
In the formula,For the orthogonal matrix of initial time,For with axis1For axis, rotates the angle θ and obtain just
Matrix is handed over,For with axis2For axis, the orthogonal matrix that the angle φ obtains is rotated.φ0, θ0The respectively pitch angle of initial time, flat
Cradle angle.
Wherein, shaft axis1And axis2Respectively by initial orthogonal matrixFirst every trade vector sum the second every trade vector
It obtains:
E3It is obtained by holder geometric properties,
Wherein,For the optical center world coordinates of initial time.
Preferably, the step of calibrating camera intrinsic parameter further include:
In the case where observing mark plate adjustment video camera pitch angle by video camera and being 0, water in video camera viewfinder is remembered
The intersection point on flat marking and sign plate both sides is a, b, and camera optical center is directed at a, b respectively, and it is flat to record state b relative status a
The difference of cradle angle be θ, from state a yawing image pick-up device to θ/2, record at this time the relative position of a and b on the image be l, then by with
Lower formula obtains focal length fxWith z-axis away from are as follows:
Wherein, the unit of zAxis is pixel, and the transforming relationship of pixel and physical size is embodied in coefficient s, the object
The unit for managing scale is millimeter.
Preferably, the step of calibrating camera intrinsic parameter further includes acceleration strategy:
Video camera is adjusted to minimum zoom value by the first step, obtains current all camera intrinsic parameters;
Video camera is directed at circular frame, measures outer parameter at this time by second step;
Third step identifies all circular frames in camera lens, calculates focal length and optical center during zoom;
Video camera yawing pitching certain angle so that circular frame is in one jiao of image, is repeated the second step by the 4th step
And third step, independent measurement go out distortion parameter.
Preferably, described that the camera intrinsic parameter and external parameters of cameras are adjusted according to presetting rule, it obtains
The step of camera intrinsic parameter and external parameters of cameras adjusted includes:
Isolated parameter is determined in the camera intrinsic parameter and external parameters of cameras;
Verify isolated parameter;
Using the isolated parameter after verifying, camera intrinsic parameter and external parameters of cameras are generated.
Disclosed herein as well is a kind of systems of sensor-based outdoor camera tracking, comprising:
Camera intrinsic parameter demarcating module is used for calibrating camera intrinsic parameter, and the camera intrinsic parameter includes focal length, light
Heart image coordinate, distortion parameter and holder parameter;
External parameters of cameras obtains module, and for obtaining external parameters of cameras, the external parameters of cameras includes orthogonal moment
Battle array and optical center world coordinates;
Projection matrix generation module, for generating projection matrix according to the camera intrinsic parameter and external parameters of cameras;
Virtual pattern broadcasts module, for broadcasting virtual pattern on caption machine according to the projection matrix.
Preferably, the system also includes:
Adjustment module is obtained for being adjusted to the camera intrinsic parameter and external parameters of cameras according to presetting rule
Obtain camera intrinsic parameter and external parameters of cameras adjusted.
Preferably, the camera intrinsic parameter demarcating module includes:
Initial intrinsic parameter acquisition submodule, for being obtained using Zhang Zhengyou algorithm in fixed camera lens zoom level and focusing
The initial intrinsic parameter of video camera under value;
Sensing data returns to submodule, for video camera to be turned to different postures, and to the mark plate of known dimensions
It takes pictures, while recording the data of lower sensor return;
Camera intrinsic parameter computational submodule, for the number according to the initial intrinsic parameter of the video camera and sensor return
According to using following formula calculating camera intrinsic parameter:
inpara=(fx,fy,u0,v0,kc1,kc2,zAxis,yAxis,φ0)
Wherein, θ, φ are respectively the yawing angle and pitch angle that sensor is passed back;xi,jFor on jth picture, i-th point
Image coordinate, PjFor the corresponding projection matrix of the corresponding video camera posture of jth picture, XiFor i-th point of world coordinates, it is
Known quantity;fx,fyFor the focal length of both direction, u0,v0For the image coordinate of both direction optical center, kc1For distortion 1, kc2For distortion
2, zAxis be z-axis away from, yAxis be y-axis away from, φ0For pitching zero deviation angle.
Compared with prior art, the application includes following advantages:
By calibrating camera intrinsic parameter, camera intrinsic parameter includes focal length, optical center image coordinate, distortion parameter and holder
Parameter, and external parameters of cameras is obtained, external parameters of cameras includes orthogonal matrix and optical center world coordinates, avoids mark plate
The problem of precision caused by area is excessive is difficult to ensure introduces more new measurement parameters, solves outdoor camera calibration
It is difficult to complete or demarcate not accurate enough problem, maintains the simplicity of Zhang Zhengyou algorithm, and improve Camera location
Precision, to be suitable for the processing and enhancing of high quality and high efficiency sports relay signal.
By being adjusted to camera intrinsic parameter and external parameters of cameras according to presetting rule, camera shooting adjusted is obtained
Machine intrinsic parameter and external parameters of cameras further improve the precision of projection matrix, to further improve Camera location
Precision.
It by acceleration strategy, avoids and repeats camera intrinsic parameter demarcating steps, solve all zoom values and gather
The problem of calibrating camera intrinsic parameter of coke number, labor intensity is reduced, reduces calibration time-consuming, improve camera intrinsic parameter
Stated accuracy, to further improve Camera location precision.
Detailed description of the invention
Fig. 1 shows a kind of step process of the embodiment of the method 1 of sensor-based outdoor camera tracking of the application
Figure;
Fig. 2 shows a kind of schematic diagrames of holder parametric geometry meaning of the application;
Fig. 3 shows the schematic diagram that a kind of video camera of the application turns to different posture shooting mark plates;
Fig. 4 shows a kind of effect diagram of Camera location of the application;
Fig. 5 shows the effect diagram of result before and after a kind of optimization problem of the application;
Fig. 6 shows a kind of schematic diagram of the circular frame for acceleration strategy of the application;
Fig. 7 shows a kind of step process of the embodiment of the method 2 of sensor-based outdoor camera tracking of the application
Figure;
Fig. 8 shows the effect diagram before and after a kind of adjustment virtual pattern of the application;
Fig. 9 shows a kind of structural block diagram of the system embodiment of sensor-based outdoor camera tracking of the application.
Specific embodiment
In order to make the above objects, features, and advantages of the present application more apparent, with reference to the accompanying drawing and it is specific real
Applying mode, the present application will be further described in detail.
One of the core concepts of the embodiments of the present invention is, firstly, calibrating camera intrinsic parameter, calculates video camera coke
Away from, optical center image coordinate, distortion parameter and holder parameter, then, obtain include orthogonal matrix and optical center world coordinates camera shooting
The outer parameter of machine, finally, projection matrix is generated using camera intrinsic parameter and external parameters of cameras, for virtual pattern in caption machine
Upper broadcast.The present invention can to avoid Surface of sign board product it is excessive caused by precision be difficult to ensure the problem of, introduce more new surveys
Parameter is measured, solves the problems, such as that camera calibration is difficult to complete or demarcate not accurate enough, the simplicity of holding Zhang Zhengyou algorithm, and
And Camera location precision is improved, to be suitable for the processing and enhancing of high quality and high efficiency sports relay signal.
Referring to Fig.1, the step of showing a kind of embodiment of the method 1 of sensor-based outdoor camera tracking of the application
Flow chart can specifically include following steps:
Step 101, calibrating camera intrinsic parameter, the camera intrinsic parameter include focal length, optical center image coordinate, distortion ginseng
Several and holder parameter.
In a particular application, camera calibration target basic object be exactly set up three-dimensional world coordinate in real space with
The mapping relations between two-dimensional pixel coordinate in image space.This mapping relations are determined by the geometrical model of video camera imaging
Fixed, determine that the process of these geometrical model relevant parameters is known as camera calibration.As concrete application of the embodiment of the present invention
A kind of example, can using national forest park in Xiaokeng as between the two-dimensional pixel coordinate in three-dimensional world coordinate and image space
Mapping relations.
Specifically, camera parameters can be subdivided into outer parameter and intrinsic parameter.Intrinsic parameter reflects video camera itself
Optical characteristics specifically includes the focal length of video camera, optical center and distortion coefficients of camera lens etc..Outer parameter reflects video camera and shooting scape
The relative positional relationship of object specifically includes the orthogonal matrix of video camera, optical center world coordinates.The inside and outside ginseng of video camera has been determined
Number also just establishes the corresponding relationship between arbitrary point in space and its imaging point on the image.Therefore, corresponding relationship
It can be described as:
Wherein, x, y, z are the world coordinates of object, are specified when establishing virtual scene, u, v, and 1 sits for homogeneous image
Mark.First matrix on the right of equationFor Intrinsic Matrix, wherein fx,fyFor the focal length of both direction, u0,v0For
The image coordinate of both direction optical center;Second matrix is by a unit matrix [I3] and a null matrix [03] composition;Third
MatrixFor outer parameter matrix, wherein variable is orthogonal matrix R33With optical center world coordinates E3, these three matrixes
Product be projection matrix.
In practical applications, it is determined, can be calculated by calibration by the characteristic of video camera itself due to the intrinsic parameter of video camera
Method passes through experiment measurement repeatedly in the lab and obtains.And the outer parameter of video camera depends on the phase of video camera and subject
It to position and posture, cannot acquire in advance, need to calculate acquisition by sensing data in actual photographed.
Since camera lens is usually constructed with distortion effect, the image coordinate obtained by above-mentioned equation also needs to carry out distortion correction ability
Obtain correct result.In a preferred embodiment of the embodiment of the present invention, calculates distortion correction and uses following formula:
u=(1+kc1r2+kc2r4)u′
v=(1+kc1r2+kc2r4)v′
Wherein,X', y', z' are coordinate of the subpoint in camera coordinates, kc1It is abnormal
Become 1, kc2For distortion 2.
Certainly, the mode of above-mentioned calculating distortion correction is intended only as example, in implementing the embodiments of the present invention, can basis
Other modes for calculating distortion correction are arranged in actual conditions, and the embodiments of the present invention are not limited thereto.In addition, in addition to above-mentioned meter
Outside the mode for calculating distortion correction, those skilled in the art can also calculate the side of distortion correction using other according to actual needs
Formula, the embodiment of the present invention are also without restriction to this.
Due to the special construction of camera pan-tilt, when video camera rotation, the position of camera optical center can also be moved.For taking
For yawing-pitching of the broadcast level of belt sensor-zoom camera machine head combination (Pan/Tilt/Zoom camera), sensor is usually returned
Yawing angle, pitch angle, zoom value, focus value numerical value.Specifically, yawing angle is that video camera CCD plane normal is flat in level
Angle of the projection relative to a reference line in the plane on face, reference line or holder factory when be set as actual zero point,
Or the projection of normal on the horizontal level is determined as relative zero when for sensor initializing.Pitch angle is defined as video camera CCD
Angle of the plane perpendicular to video camera CCD plane long side upward direction and certain reference line, reference line or be the flat of actual zero point
Rocker, or be sensor initializing when, relative zero of the video camera CCD plane perpendicular to video camera CCD long side upward direction.Become
Coke number and focus value generally return to integer, and range is the maximum value from zero to one, which presets when holder dispatches from the factory.
Wherein, video camera CCD plane is the plane of delineation above-mentioned.
Referring to Fig. 2, a kind of schematic diagram of holder parametric geometry meaning of the application is shown.As shown in Fig. 2, Tilt Axis
For pitch axis, Pan Axis is yawing axis, and straight line with the arrow is optical axis, and it is y-axis away from, φ that zAxis, which is z-axis away from, yAxis,0For
Pitching zero deviation angle.
For Camera location, holder itself can affect the additional parameter of Camera location precision.For quality
For higher holder, it can be described with three parameters.Three parameters specifically can be respectively y-axis away from, z-axis away from and pitching
Zero deviation angle φ0.Wherein, y-axis is away from the distance for optical center to pitch axis;Z-axis away from for when video camera CCD plane peace rocker it is flat
When row, pitch axis is total to vertical line at a distance from the intersection point to optical center on optical axis with optical axis;Pitching zero deviation angle φ0To work as sensor
When pitch angle is 0, the angle of yawing axis and video camera CCD plane upward direction.If holder is the holder for having actual zero point, bow
Facing upward zero deviation angle should be 0, in fact, mismachining tolerance, which will lead to the value, deviates zero.
Accordingly, it is determined that the problem of projection matrix, can specifically describe are as follows:
First, predefine camera intrinsic parameter matrix, i.e. focal length fx、fy, optical center position u0、v0, distortion parameter kc1、
kc2And holder parameter, including z-axis away from, y-axis away from pitching zero deviation angle φ0.Focal length, optical center position and distortion parameter are equal
Can be the variable of zoom value and focus value, thus answer precalculated focal length, optical center position and distortion parameter and zoom value,
The relationship and z-axis of focus value away from, y-axis away from and pitching zero deviation angle φ0The relationship of peaceful cradle angle, pitch angle.
Second, in practical application, the zoom value, the focus value that are returned by sensor, in precalculated focal length, optical center
Correspondingly focal length, optical center position and distortion parameter are searched in the relationship of position and distortion parameter and zoom value and focus value.
And returned by sensor yawing angle, pitch angle, precalculate z-axis away from, y-axis away from pitching zero deviation angle φ0Peace
Cradle angle, pitch angle relationship in search correspondingly z-axis away from, y-axis away from pitching zero deviation angle φ0, to calculate the throwing at per moment
Shadow matrix.
Referring to Fig. 3, the schematic diagram that a kind of video camera of the application turns to different posture shooting mark plates is shown.Such as Fig. 3
Shown, fixed signal Board position is motionless, and rotation video camera to posture shoots mark plate.
In one preferred embodiment of the invention, the step 101 can specifically include following sub-step:
Sub-step S11 is obtained at the beginning of the video camera under fixed camera lens zoom level and focus value using Zhang Zhengyou algorithm
Beginning intrinsic parameter;
Video camera is turned to different postures, and taken pictures to the mark plate of known dimensions by sub-step S12, is remembered simultaneously
Record the data that lower sensor returns;
Sub-step S13 is calculated according to the data that the initial intrinsic parameter of the video camera and sensor return using following formula
Camera intrinsic parameter:
inpara=(fx,fy,u0,v0,kc1,kc2,zAxis,yAxis,φ0)
Wherein, θ, φ are respectively the yawing angle and pitch angle that sensor is passed back;xi,jFor on jth picture, i-th point
Image coordinate, PjFor the corresponding projection matrix of the corresponding video camera posture of jth picture, XiFor i-th point of world coordinates, it is
Known quantity;fx,fyFor the focal length of both direction, u0,v0For the image coordinate of both direction optical center, kc1For distortion 1, kc2For distortion
2, zAxis be z-axis away from, yAxis be y-axis away from, φ0For pitching zero deviation angle.
In the concrete realization, camera calibration problem is computer vision field root problem.But it is applied to outdoor TV
In the problem of being broadcast live, being particularly applied to sports event broadcast, there is two o'clock outstanding request.First, it is desirable that the side of calibrating camera
Method Camera location precision with higher, enables virtual pattern to fit closely outdoor scene always, as shown in Figure 4;Second, it is desirable that mark
The method for determining video camera can apply to the outdoor large scene that focal distance is tens meters.
Camera calibration problem is the problem of determining video camera basic parameter by one group of experiment.It needs to be determined that video camera base
This parameter is focal length of camera, optical center position, distortion parameter and holder parameter, including z-axis away from, y-axis away from inclined with pitching zero point
Difference angle phi0.In one preferred embodiment of the invention, experimental method is determined as image method, i.e., by shooting to video camera
One group of image analyzed, obtain video camera basic parameter.
Firstly, fixed camera lens zoom level and focus value, and record zoom value and focus value at this time.Utilize Zhang Zhengyou
Classic algorithm obtain camera intrinsic parameter initial value.Then video camera is turned into different postures, and to known dimensions
Mark plate is taken pictures, while recording the data of lower sensor return.Then optimization problem can be described as:
inpara=(fx,fy,u0,v0,kc1,kc2,zAxis,yAxis,φ0)
Certainly, the processing method of above-mentioned calibrating camera intrinsic parameter is intended only as example, in implementing the embodiments of the present invention,
The processing method of other calibrating camera intrinsic parameters can be set according to the actual situation, and the embodiment of the present invention does not limit this
System.In addition, those skilled in the art can also be according to actual needs other than the processing method of above-mentioned calibrating camera intrinsic parameter
Using the processing method of other calibrating camera intrinsic parameters, the embodiment of the present invention is also without restriction to this.
Referring to Fig. 5, the effect diagram of result before and after a kind of optimization problem of the application is shown.As shown in figure 5, optimization is asked
The lattice point difference that topic front and back is identified in image left and right ends is larger.
Due to focal length fxAnd z-axis is degeneracy away between, the optimal solution of optimization problem is not a point in high latitude space, and
It is a hypercurve.
Therefore, in one preferred embodiment of the invention, the step 101 specifically can also include:
In the case where observing mark plate adjustment video camera pitch angle by video camera and being 0, water in video camera viewfinder is remembered
The intersection point on flat marking and sign plate both sides is a, b, and camera optical center is directed at a, b respectively, and it is flat to record state b relative status a
The difference of cradle angle be θ, from state a yawing image pick-up device to θ/2, record at this time the relative position of a and b on the image be l, then by with
Lower formula obtains focal length fxWith z-axis away from are as follows:
Wherein, the unit of zAxis is pixel, and the transforming relationship of pixel and physical size is embodied in coefficient s, the object
The unit for managing scale is millimeter.
In the concrete realization, for the combination of any zoom value and focus value, all combinations can pass through the above method
Calculate the relationship of camera intrinsic parameter and zoom value, the relationship of focus value and holder parameter peace cradle angle, pitch angle.
Referring to Fig. 6, a kind of schematic diagram of the circular frame for acceleration strategy of the application is shown.
Since focus value usually changes smaller in practical applications, for the measurement of focus value, grouping mea-sure can be used,
The method of linear interpolation.However zoom value can change greatly in practical applications.
Therefore, in one preferred embodiment of the invention, the step 101 can also include specifically acceleration strategy:
Video camera is adjusted to minimum zoom value by sub-step S21, obtains current all camera intrinsic parameters;
Video camera is directed at circular frame, measures outer parameter at this time by sub-step S22;
Sub-step S23 identifies all circular frames in camera lens, calculates focal length and optical center during zoom;
Video camera yawing pitching certain angle so that circular frame is in one jiao of image, is repeated the son by sub-step S24
Step S22 and sub-step S23, independent measurement go out distortion parameter.
In the concrete realization, intrinsic parameter when video camera minimum zoom value can be demarcated using aforementioned algorism, will be taken the photograph
Camera is directed at circular frame, and circular frame is made to be located at camera center position, measures outer parameter at this time, due in video camera zoom,
Outer parameter constant, image change is caused by intrinsic parameter variation, therefore during zoom, is identified all back-shaped in camera lens
Frame can calculate focal length and optical center, and video camera yawing pitching certain angle so that circular frame is in one jiao of image, repeats
The sub-step S22 and sub-step S23 can go out distortion parameter with independent measurement.
It by acceleration strategy, avoids and repeats camera intrinsic parameter demarcating steps, solve all zoom values and gather
The problem of calibrating camera intrinsic parameter of coke number, labor intensity is reduced, reduces calibration time-consuming, improve camera intrinsic parameter
Stated accuracy, to further improve Camera location precision.
Step 102, external parameters of cameras is obtained, the external parameters of cameras includes orthogonal matrix and optical center world coordinates.
In practical applications, by yawing angle θ, pitch angle φ and three holder parameters zAxis, yAxis, φ0Obtain outer ginseng
Matrix number.
In one preferred embodiment of the invention, the step 102 can specifically include following sub-step:
Sub-step S301 calculates outer parameter matrix using following formula according to the holder parameter:
Wherein, orthogonal matrix R33It is obtained by following formula:
In the formula,For the orthogonal matrix of initial time,For with axis1For axis, rotates the angle θ and obtain just
Matrix is handed over,For with axis2For axis, the orthogonal matrix that the angle φ obtains is rotated.φ0, θ0The respectively pitch angle of initial time, flat
Cradle angle.
Wherein, shaft axis1And axis2Respectively by initial orthogonal matrixFirst every trade vector sum the second every trade vector
It obtains:
E3It is obtained by holder geometric properties,
Wherein,For the optical center world coordinates of initial time.
In a particular application, orthogonal matrix R is calculated33The pitching of Shi Bixu elder generation, rear yawing.
Certainly, the processing method of above-mentioned acquisition external parameters of cameras is intended only as example, in implementing the embodiments of the present invention,
Other processing methods for obtaining external parameters of cameras can be set according to the actual situation, and the embodiment of the present invention does not limit this
System.In addition, those skilled in the art can also be according to actual needs other than the processing method of above-mentioned acquisition external parameters of cameras
Using other processing methods for obtaining external parameters of cameras, the embodiment of the present invention is also without restriction to this.
By calibrating camera intrinsic parameter, camera intrinsic parameter includes focal length, optical center image coordinate, distortion parameter and holder
Parameter, and external parameters of cameras is obtained, external parameters of cameras includes orthogonal matrix and optical center world coordinates, avoids mark plate
The problem of precision caused by area is excessive is difficult to ensure introduces more new measurement parameters, solves outdoor camera calibration
It is difficult to complete or demarcate not accurate enough problem, maintains the simplicity of Zhang Zhengyou algorithm, and improve Camera location
Precision, to be suitable for the processing and enhancing of high quality and high efficiency sports relay signal.
Step 103, projection matrix is generated according to the camera intrinsic parameter and external parameters of cameras.
Specifically, by camera intrinsic parameter fx, fy, u0, v0, kc1, kcx substitutionConstitute intrinsic parameter square
Battle array, by external parameters of cameras R33, E3It substitutes intoOuter parameter matrix is constituted, to obtain projection matrix and be
Certainly, the mode of above-mentioned calculating projection matrix is intended only as example, in implementing the embodiments of the present invention, can basis
Other modes for calculating projection matrix are arranged in actual conditions, and the embodiments of the present invention are not limited thereto.In addition, in addition to above-mentioned meter
Outside the mode for calculating projection matrix, those skilled in the art can also calculate the side of projection matrix using other according to actual needs
Formula, the embodiment of the present invention are also without restriction to this.
Step 104, according to the projection matrix, virtual pattern is broadcasted on caption machine.
In the concrete realization, caption machine is usually to be made of PC plus the subtitle superposition card of profession and corresponding software, is used
In stacking chart's text curtain in vision signal.Its feature is: in real time, bright in luster without generation, the loss of signal is small for superposition, fits
Logo, footmark, winged advertisement in a left side etc. are superimposed together in TV station.
Caption machine is made of computer, title insert and subtitle Software for producing three parts.Present computer is very universal, greatly
Family is also very familiar with it;Subtitle Software for producing is that control subtitle is added on background frame, and the tool interacted with user does not have
Subtitle software, user can not just control computer, can not just make subtitle;After having computer and subtitle software,
Must have one can allow for background frame to input, and after superimposed subtitle, and final output is set with credit video
Standby, title insert is exactly such equipment.
In the system setting of subtitle software, it can be seen that have coding mode and bypass mode, bypass mode is exactly subtitle
Card does not work, and coding mode is exactly title insert work.Because title insert will carry out modulus to the background video of input at work
Then conversion carries out digital-to-analogue conversion when it is exported, so, when system is set as coding mode, background video will be by
The influence converted to modulus, digital-to-analogue.When system is set as bypass mode, vision signal does not have any conversion, is only inputting
Delivery outlet forms loop and is not lost theoretically for coding mode.
After system is set as coding mode, next there are three option, first is that subtitle superposition, second is that background is straight
Lead to, third is that caption frame is deposited;Title insert is divided into two layers, and one layer is background video, and one layer is subtitle.When subtitle superposition in choosing, word
Curtain software will control the aliasing device in title insert, the subtitle made is aliased on background frame, the video at this moment exported
It has been just the video for folding subtitle;When background is straight-through in choosing, aliasing device does not work, and subtitle layer is not done mutually with background layer
It relates to, background video only makees simple modulus, digital-to-analogue conversion, then exports, and output is the video of not superimposed subtitle;In elected
When caption frame is deposited, aliasing device does not also work, and the background video that subtitle card docking receives not exports, and only exports subtitle layer, so
The only subtitle of final output, without background video.
Referring to Fig. 7, the step of showing a kind of embodiment of the method 2 of sensor-based outdoor camera tracking of the application
Flow chart can specifically include following steps:
Step 201, calibrating camera intrinsic parameter, the camera intrinsic parameter include focal length, optical center image coordinate, distortion ginseng
Several and holder parameter.
In the concrete realization, those skilled in the art are according to the actual situation using the classic algorithm mark that Zhang Zhengyou can be used
Determine camera intrinsic parameter, the other algorithm calibrating camera internal references of acceleration strategy calibrating camera intrinsic parameter etc. also can be used
Number, the application is to this without limiting.
Step 202, external parameters of cameras is obtained, the external parameters of cameras includes orthogonal matrix and optical center world coordinates.
It in practical applications, can be by three parameters zAxis, yAxis, φ of yawing angle θ, pitch angle φ and holder0
Obtain outer parameter matrix.
Step 203, the camera intrinsic parameter and external parameters of cameras are adjusted according to presetting rule, are adjusted
Camera intrinsic parameter and external parameters of cameras afterwards.
In the concrete realization, particularly, in outdoor application video camera, the problem of still facing error.Error is come
Source mainly has two:
First is that the problem of mark plate itself makes.The problems such as mark plate of production has a flexure, fold causes lattice point real
Border world coordinates is not inconsistent with ideal world coordinate.
Second is that object cannot generate clear image in video camera observation ward when focus value is suitable for outdoor.Lead to lattice
Point image measurement of coordinates inaccuracy.
The main foundation that adjustable strategies are formulated is isolated parameter, that is, finds a phenomenon so that it the sum of as far as possible some
Parameter is related.
Referring to Fig. 8, the effect diagram before and after a kind of adjustment virtual pattern of the application is shown.As shown in figure 8, white
Virtual pattern between actual graphical is not matched that with actual graphical.
In one preferred embodiment of the invention, the step 203 can specifically include following sub-step:
Sub-step S401 determines isolated parameter in the camera intrinsic parameter and external parameters of cameras;
Sub-step S402 verifies isolated parameter;
Sub-step S403 generates camera intrinsic parameter and external parameters of cameras using the isolated parameter after verifying.
In practical applications, although focal length fxWith z-axis away from the position that can all influence virtual pattern, but focal length fxIt can be with shadow
Ring the size of virtual pattern.Therefore, after substantially adjusting virtual pattern position, so that it may by according to the big of virtual pattern
The size of small and practical reference pattern determines focal length fx。
Certainly, the processing method of above-mentioned adjustment virtual pattern is intended only as example, in implementing the embodiments of the present invention, can be with
The processing method of other adjustment virtual patterns is set according to the actual situation, and the embodiments of the present invention are not limited thereto.In addition, removing
Outside the processing method of above-mentioned adjustment virtual pattern, those skilled in the art can also be according to actual needs using other adjustment void
The processing method of quasi- figure, the embodiment of the present invention are also without restriction to this.
Step 204, projection matrix is generated according to the camera intrinsic parameter and external parameters of cameras.
In a particular application, projection matrix is
Step 205, according to the projection matrix, virtual pattern is broadcasted on caption machine.
In the concrete realization, those skilled in the art formulate different shape, the virtual graph that difference needs according to the actual situation
Shape, such as swimming lane number, swimming lane number corresponding name and ranking, world record line etc. of swimming event match, the application
To this without limiting.
To make those skilled in the art more fully understand the application, below by taking video camera indoor standardization, outdoor adjustment as an example,
The embodiment of the present invention will be further explained.
For Camera location, including indoor camera tracking and outdoor camera tracking, so-called indoor or outdoors
Refer to subject to the distance of video camera, i.e. object distance.When the distance of subject to video camera and focus value institute are right
When the object distance answered is closer to, subject could blur-free imaging.Outdoor camera tracking, focus value generally require to adjust
To five meters or more, i.e., for object distance often at five meters or more, however when object distance is larger, subject is smaller in the picture, if
Camera calibration is carried out using mark plate, then the size of mark plate needs very big.
Focus value is adjusted to suitable for outdoor value by the first step.
Specifically, being suitable for outdoor value is video camera in outside work, subject at a distance from video camera,
Using the distance as video camera focus value.
Aperture is turned down to mark plate lattice point not gone out by second step.
It should be noted that the degree that aperture can be adjusted depends on illumination condition, aperture is smaller, and the depth of field is bigger, thus
Fog-level caused by can reducing because of image out-focus.
Third step measures camera intrinsic parameter.
As an example, the measurement camera intrinsic parameter can be using in the classic algorithm calibrating camera of Zhang Zhengyou
Parameter.
4th step finely tunes camera parameters, obtains best Camera location effect.
In a particular application, when demarcating the camera parameters of outside work indoors, the image of shooting is unavoidably mould
Paste.Error caused by being obscured as image will be passed to camera parameters, and Camera location precision will reduce, therefore adjustment is
It is necessary.
For embodiment of the method, for simple description, therefore, it is stated as a series of action combinations, but this field
Technical staff should be aware of, and embodiment of that present invention are not limited by the describe sequence of actions, because implementing according to the present invention
Example, some steps may be performed in other sequences or simultaneously.Secondly, those skilled in the art should also know that, specification
Described in embodiment belong to preferred embodiment, the actions involved are not necessarily necessary for embodiments of the present invention.
With reference to Fig. 9, a kind of structure of the system embodiment of sensor-based outdoor camera tracking of the application is shown
Block diagram can specifically include with lower module:
Camera intrinsic parameter demarcating module 901, is used for calibrating camera intrinsic parameter, and the camera intrinsic parameter may include
Focal length, optical center image coordinate, distortion parameter and holder parameter;
External parameters of cameras obtains module 902, and for obtaining external parameters of cameras, the external parameters of cameras may include
Orthogonal matrix and optical center world coordinates;
Projection matrix generation module 903, for generating projection square according to the camera intrinsic parameter and external parameters of cameras
Battle array;
Virtual pattern broadcasts module 904, for that can broadcast virtual pattern on caption machine according to the projection matrix.
As a kind of preferred embodiment, the system also includes:
Module 905 is adjusted, for being adjusted to the camera intrinsic parameter and external parameters of cameras according to presetting rule,
Obtain camera intrinsic parameter and external parameters of cameras adjusted.
As a kind of preferred embodiment, the camera intrinsic parameter demarcating module 901 includes:
Initial intrinsic parameter acquisition submodule, for being obtained using Zhang Zhengyou algorithm in fixed camera lens zoom level and focusing
The initial intrinsic parameter of video camera under value;
Sensing data returns to submodule, for video camera to be turned to different postures, and can be to the mark of known dimensions
Will plate is taken pictures, while recording the data of lower sensor return;
Camera intrinsic parameter computational submodule, for the number according to the initial intrinsic parameter of the video camera and sensor return
According to using following formula calculating camera intrinsic parameter:
inpara=(fx,fy,u0,v0,kc1,kc2,zAxis,yAxis,φ0)
Wherein, θ, φ can be respectively the yawing angle and pitch angle that sensor is passed back;xi,jCan on jth picture, the
The image coordinate of i point, PjIt can be the corresponding projection matrix of the corresponding video camera posture of jth picture, XiIt can be i-th
The world coordinates of point is known quantity;fx,fyIt can be the focal length of both direction, u0,v0It can be the image of both direction optical center
Coordinate, kc1It can be distortion 1, kc2It can be 2, zAxis of distortion can be z-axis away from, yAxis can be y-axis away from, φ0Can be
Pitching zero deviation angle.
As a kind of preferred embodiment, the external parameters of cameras obtains module 902 and includes:
Outer parameter matrix computational submodule, for calculating outer parameter matrix using following formula according to the holder parameter:
Wherein, orthogonal matrix R33It can be obtained by following formula:
In the formula,It can be the orthogonal matrix of initial time,It can be for axis1For axis, the angle θ is rotated
The orthogonal matrix of acquisition,It can be for axis2For axis, the orthogonal matrix that the angle φ obtains is rotated.φ0, θ0It can be respectively just
The pitch angle at moment beginning, yawing angle.
Wherein, shaft axis1And axis2It can be respectively by initial orthogonal matrixFirst the second every trade of every trade vector sum
Vector obtains:
E3It can be obtained by holder geometric properties,
Wherein,It can be the optical center world coordinates of initial time.
As a kind of preferred embodiment, the camera intrinsic parameter demarcating module 901 further include:
Degeneracy releases submodule, for observing mark plate by video camera, adjusts the case where video camera pitch angle is 0
Under, remember that the intersection point on horizontal marking and sign plate both sides in video camera viewfinder is a, b, camera optical center is directed at a, b respectively, is remembered
The difference for recording lower state b relative status a yawing angle is θ, and from state a yawing image pick-up device to θ/2, a and b is on the image at this time for record
Relative position is l, then obtains focal length f by following formulaxWith z-axis away from are as follows:
Wherein, the unit of zAxis can be pixel, and the transforming relationship of pixel and physical size is embodied in coefficient s, institute
The unit for stating physical size can be millimeter.
As a kind of preferred embodiment, the camera intrinsic parameter demarcating module 901 further include:
Current intrinsic parameter acquisition submodule obtains current all video cameras for video camera to be adjusted to minimum zoom value
Intrinsic parameter;
Outer parameter measurement submodule measures outer parameter at this time for video camera to be directed at circular frame;
Focal length and optical center computational submodule are calculated for identifying all circular frames in camera lens during zoom
Discharge of the coke away from and optical center;
Distortion parameter measures submodule, is used for by video camera yawing pitching certain angle, so that circular frame is in image one
Angle, repeats the second step and third step, independent measurement go out distortion parameter.
As a kind of preferred embodiment, the adjustment module 905 includes:
Isolated parameter determination submodule, for determining isolated parameter in the camera intrinsic parameter and external parameters of cameras;
Isolated Verification submodule, for verifying isolated parameter;
Parameter generates submodule, for generating and joining outside camera intrinsic parameter and video camera using the isolated parameter after verifying
Number.
For system embodiments, since it is basically similar to the method embodiment, related so being described relatively simple
Place illustrates referring to the part of embodiment of the method.
All the embodiments in this specification are described in a progressive manner, the highlights of each of the examples are with
The difference of other embodiments, the same or similar parts between the embodiments can be referred to each other.
Above to a kind of method and system of sensor-based outdoor camera tracking provided herein, carry out
It is discussed in detail, specific examples are used herein to illustrate the principle and implementation manner of the present application, above embodiments
Illustrate to be merely used to help understand the present processes and its core concept;At the same time, for those skilled in the art, according to
According to the thought of the application, there will be changes in the specific implementation manner and application range, in conclusion the content of the present specification
It should not be construed as the limitation to the application.
Claims (7)
1. a kind of method of sensor-based outdoor camera tracking characterized by comprising
Calibrating camera intrinsic parameter, the camera intrinsic parameter include focal length, optical center image coordinate, distortion parameter and holder ginseng
Number;
External parameters of cameras is obtained, the external parameters of cameras includes orthogonal matrix and optical center world coordinates;
Projection matrix is generated according to the camera intrinsic parameter and external parameters of cameras;
According to the projection matrix, virtual pattern is broadcasted on caption machine;
Wherein, the step of calibrating camera intrinsic parameter includes:
The initial intrinsic parameter of video camera under fixed camera lens zoom level and focus value is obtained using Zhang Zhengyou algorithm;
Video camera is turned into different postures, and is taken pictures to the mark plate of known dimensions, while recording lower sensor return
Data;
According to the data that the initial intrinsic parameter of the video camera and sensor return, camera intrinsic parameter is calculated using following formula:
Inpara=(fx,fy,u0,v0,kc1,kc2,zAxis,yAxis,φ0)
Wherein, θ, φ are respectively the yawing angle and pitch angle that sensor is passed back;xi,jFor on jth picture, i-th point of image
Coordinate, PjFor the corresponding projection matrix of the corresponding video camera posture of jth picture, XiIt is known for i-th point of world coordinates
Amount;fx,fyFor the focal length of both direction, u0,v0For the image coordinate of both direction optical center, kc1For distortion 1, kc2For distortion 2,
It is y-axis away from, φ that zAxis, which is z-axis away from, yAxis,0For pitching zero deviation angle.
2. the method according to claim 1, wherein it is described according to the camera intrinsic parameter and video camera outside
Parameter generated before the step of projection matrix, further includes:
The camera intrinsic parameter and external parameters of cameras are adjusted according to presetting rule, obtained in video camera adjusted
Parameter and external parameters of cameras.
3. the method according to claim 1, wherein the holder parameter include z-axis away from zAxis, y-axis away from
YAxis and pitching zero deviation angle φ0, the step of acquisition external parameters of cameras includes:
Outer parameter matrix is calculated using following formula according to the holder parameter:
Wherein, orthogonal matrix R33It is obtained by following formula:
In the formula,For the orthogonal matrix of initial time,For with axis1For axis, the orthogonal moment that the angle θ obtains is rotated
Battle array,For with axis2For axis, the orthogonal matrix that the angle φ obtains, φ are rotated0, θ0The respectively pitch angle of initial time, yawing
Angle,For null matrix [03] turn battle array;
Wherein, shaft axis1And axis2Respectively by initial orthogonal matrixFirst every trade vector sum the second every trade vector obtain:
E3It is obtained by holder geometric properties,
Wherein, E3For optical center world coordinates;For the optical center world coordinates of initial time.
4. according to the method described in claim 3, it is characterized in that, the step of the calibrating camera intrinsic parameter further include:
In the case where observing mark plate adjustment video camera pitch angle by video camera and being 0, horizontal mark in video camera viewfinder is remembered
Line and the intersection point on mark plate both sides are a, b, and camera optical center is directed at a, b respectively, records state b relative status a yawing angle
Difference be θ, from state a yawing image pick-up device to θ/2, record at this time the relative position of a and b on the image be l, then pass through following public affairs
Formula obtains focal length fxWith z-axis away from are as follows:
Wherein, the unit of zAxis is pixel, and in the transforming relationship of pixel and physical size, the unit of the physical size is milli
Rice.
5. method according to claim 1 or 4, which is characterized in that the step of calibrating camera intrinsic parameter further includes
Acceleration strategy:
Video camera is adjusted to minimum zoom value by the first step, obtains current all camera intrinsic parameters;
Video camera is directed at circular frame, measures outer parameter at this time by second step;
Third step identifies all circular frames in camera lens, calculates focal length and optical center during zoom;
Video camera yawing pitching certain angle so that circular frame is in one jiao of image, is repeated the second step and the by the 4th step
Three steps, independent measurement go out distortion parameter.
6. a kind of system of sensor-based outdoor camera tracking characterized by comprising
Camera intrinsic parameter demarcating module is used for calibrating camera intrinsic parameter, and the camera intrinsic parameter includes focal length, optical center figure
As coordinate, distortion parameter and holder parameter;
External parameters of cameras obtains module, for obtaining external parameters of cameras, the external parameters of cameras include orthogonal matrix and
Optical center world coordinates;
Projection matrix generation module, for generating projection matrix according to the camera intrinsic parameter and external parameters of cameras;
Virtual pattern broadcasts module, for broadcasting virtual pattern on caption machine according to the projection matrix;
Wherein, the camera intrinsic parameter demarcating module includes:
Initial intrinsic parameter acquisition submodule, for being obtained under fixed camera lens zoom level and focus value using Zhang Zhengyou algorithm
The initial intrinsic parameter of video camera;
Sensing data returns to submodule, for video camera to be turned to different postures, and carries out to the mark plate of known dimensions
It takes pictures, while recording the data of lower sensor return;
Camera intrinsic parameter computational submodule is adopted for the data according to the initial intrinsic parameter of the video camera and sensor return
Camera intrinsic parameter is calculated with following formula:
Inpara=(fx,fy,u0,v0,kc1,kc2,zAxis,yAxis,φ0)
Wherein, θ, φ are respectively the yawing angle and pitch angle that sensor is passed back;xi,jFor on jth picture, i-th point of image
Coordinate, PjFor the corresponding projection matrix of the corresponding video camera posture of jth picture, XiIt is known for i-th point of world coordinates
Amount;fx,fyFor the focal length of both direction, u0,v0For the image coordinate of both direction optical center, kc1For distortion 1, kc2For distortion 2,
It is y-axis away from, φ that zAxis, which is z-axis away from, yAxis,0For pitching zero deviation angle.
7. system according to claim 6, which is characterized in that the system also includes:
Adjustment module is adjusted for being adjusted to the camera intrinsic parameter and external parameters of cameras according to presetting rule
Camera intrinsic parameter and external parameters of cameras after whole.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310596129.9A CN103729839B (en) | 2013-11-21 | 2013-11-21 | A kind of method and system of sensor-based outdoor camera tracking |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310596129.9A CN103729839B (en) | 2013-11-21 | 2013-11-21 | A kind of method and system of sensor-based outdoor camera tracking |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103729839A CN103729839A (en) | 2014-04-16 |
CN103729839B true CN103729839B (en) | 2018-12-25 |
Family
ID=50453900
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310596129.9A Expired - Fee Related CN103729839B (en) | 2013-11-21 | 2013-11-21 | A kind of method and system of sensor-based outdoor camera tracking |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103729839B (en) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105516661B (en) * | 2015-12-10 | 2019-03-29 | 吴健辉 | Principal and subordinate's target monitoring method that fisheye camera is combined with ptz camera |
CN107610185A (en) * | 2017-10-12 | 2018-01-19 | 长沙全度影像科技有限公司 | A kind of fisheye camera fast calibration device and scaling method |
CN109982005A (en) * | 2017-12-27 | 2019-07-05 | 艾迪普(北京)文化科技股份有限公司 | A kind of picture and text based on panoramic video follow synthetic method |
CN108805940B (en) * | 2018-06-27 | 2021-06-04 | 亿嘉和科技股份有限公司 | Method for tracking and positioning zoom camera in zooming process |
CN112272272B (en) * | 2020-11-30 | 2022-03-22 | 成都中信华瑞科技有限公司 | Imaging method and device |
CN112348493A (en) * | 2021-01-07 | 2021-02-09 | 北京电信易通信息技术股份有限公司 | Intelligent conference recording system and method |
CN114758011B (en) * | 2022-04-13 | 2023-02-17 | 南京航空航天大学 | Zoom camera online calibration method fusing offline calibration results |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101179500A (en) * | 2007-10-30 | 2008-05-14 | 北京航空航天大学 | Method for implementing enhancement type video service mode of access gridding |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0437964A (en) * | 1990-06-01 | 1992-02-07 | Eastman Kodak Japan Kk | Document reading device |
-
2013
- 2013-11-21 CN CN201310596129.9A patent/CN103729839B/en not_active Expired - Fee Related
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101179500A (en) * | 2007-10-30 | 2008-05-14 | 北京航空航天大学 | Method for implementing enhancement type video service mode of access gridding |
Non-Patent Citations (2)
Title |
---|
电视直播虚拟广告系统关键技术的研究;殷伟良;《中国优秀硕士论文全文数据库》;20130715;摘要,第50-58页 * |
计算机立体视觉中的摄像机视觉标定算法的研究与应用;黄侃等;《中国优秀硕士论文全文数据库》;20071115;摘要,正文第7-46页 * |
Also Published As
Publication number | Publication date |
---|---|
CN103729839A (en) | 2014-04-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103729839B (en) | A kind of method and system of sensor-based outdoor camera tracking | |
CN105118055B (en) | Camera position amendment scaling method and system | |
US9286680B1 (en) | Computational multi-camera adjustment for smooth view switching and zooming | |
CN109146794B (en) | A kind of light field image rotation error bearing calibration | |
TWI555379B (en) | An image calibrating, composing and depth rebuilding method of a panoramic fish-eye camera and a system thereof | |
CN105488810B (en) | A kind of focusing light-field camera inside and outside parameter scaling method | |
US20110026014A1 (en) | Methods and systems for calibrating an adjustable lens | |
US20180213218A1 (en) | Equipment and method for promptly performing calibration and verification of intrinsic and extrinsic parameters of a plurality of image capturing elements installed on electronic device | |
CN110782394A (en) | Panoramic video rapid splicing method and system | |
KR101342393B1 (en) | Georeferencing Method of Indoor Omni-Directional Images Acquired by Rotating Line Camera | |
CN109615663A (en) | Panoramic video bearing calibration and terminal | |
Sinha et al. | Towards calibrating a pan-tilt-zoom camera network | |
WO2014023231A1 (en) | Wide-view-field ultrahigh-resolution optical imaging system and method | |
JP2009284188A (en) | Color imaging apparatus | |
CN111461963B (en) | Fisheye image stitching method and device | |
CN107302668A (en) | High dynamic range imaging module based on runner dynamic light splitting | |
CN102589529B (en) | Scanning close-range photogrammetry method | |
CN106447735A (en) | Panoramic camera geometric calibration processing method | |
CN102098442A (en) | Method and system for calibrating non-overlap ratio of optical axis and visual axis of zoom camera | |
CN107274352A (en) | A kind of image processing method and the real-time sampling system applied to lens distortion and photography distortion correction | |
Ikeda et al. | Calibration method for an omnidirectional multicamera system | |
CN111698467B (en) | Intelligent tracking method and system based on multiple cameras | |
CN103546680B (en) | A kind of deformation-free omni-directional fisheye photographic device and a method for implementing the same | |
CN102497511B (en) | Method for shooting object moving at high speed by using low-speed camera | |
JP2020194454A (en) | Image processing device and image processing method, program, and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20190321 Address after: 100195 No. 621, 6th floor, No. 1 Building, 131 North West Fourth Ring Road, Haidian District, Beijing Patentee after: BEIJING XIN'AOTE WISDOM SPORTS INNOVATION DEVELOPMENT Co.,Ltd. Address before: 100195 Haidian District Wukesong 49 new science and technology building, Beijing Patentee before: China Digital Video (Beijing) Co.,Ltd. |
|
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20181225 |