CN107330934A - The boundling Adjustable calculation method and system of low dimensional - Google Patents
The boundling Adjustable calculation method and system of low dimensional Download PDFInfo
- Publication number
- CN107330934A CN107330934A CN201710370360.4A CN201710370360A CN107330934A CN 107330934 A CN107330934 A CN 107330934A CN 201710370360 A CN201710370360 A CN 201710370360A CN 107330934 A CN107330934 A CN 107330934A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msubsup
- msub
- jth
- view
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
- G06T7/579—Depth or shape recovery from multiple images from motion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/285—Analysis of motion using a sequence of stereo image pairs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
- G06T7/337—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving reference images or patches
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Medicines Containing Antibodies Or Antigens For Use As Internal Diagnostic Agents (AREA)
Abstract
The invention provides a kind of boundling Adjustable calculation method and system of low dimensional, including：Determine the initial value of kinematic parameter；Calculating is optimized to the object function of kinematic parameter, the kinematic parameter after being optimized；According to the kinematic parameter after optimization, three-dimensional scenic point coordinates is calculated.The depth of field of several views is expressed as the function of the relative movement parameters of view two-by-two by the present invention, realize from several views and directly recover kinematic parameter, parsing obtains three-dimensional scenic point coordinates from kinematic parameter again, so as to be rejected in the parameter optimisation procedure that three-dimensional scenic point coordinates is adjusted from boundling, the dimension of parameter space is significantly reduced.The present invention be it is a kind of initialize that easy, Shandong nation property is good, calculating speed faster, the higher low dimensional collection beam adjusting method of computational accuracy.The present invention can be used as the core calculations engine of the applications such as unmanned vehicle/unmanned plane vision guided navigation, 3 D visual reconstruction, augmented reality.
Description
Technical field
The present invention relates to computer vision, photogrammetric field, meter is adjusted in particular to a kind of boundling of low dimensional
Calculate method and system.
Background technology
Boundling adjusts (Bundle Adjustment), i.e., recover three-dimensional scenic point coordinates, kinematic parameter from several views
And camera parameter, it is computer vision and one of the core technology in field such as photogrammetric.The target of boundling adjustment technology is to make
The re-projection error for obtaining picture point is minimized, and re-projection error is represented by three-dimensional scenic point coordinates, kinematic parameter and camera
Nonlinearity in parameters function.For there is the situation at m three dimensional field sight spot and n width views, parameter space is tieed up for 3*m+6*n.Due to
The number at three dimensional field sight spot is generally very big, causes the dimension of parameter space to be optimized huge.At present, the main flow side of boundling adjustment
Method is the nonlinear optimization algorithm realization openness using parameter Jacobi (Jacobian) matrix is considered, speed is calculated to be lifted
Degree, but the dimension of the parameter space of main stream approach is more, still needs further improvement, to adapt to the demand calculated in real time.
The content of the invention
For defect of the prior art, it is an object of the invention to provide a kind of boundling Adjustable calculation method of low dimensional with
System.The depth of field of several views is expressed as the function of the relative movement parameters of view two-by-two by the present invention, realizes and is regarded from several
Figure directly recovers kinematic parameter, then parsing obtains three-dimensional scenic point coordinates from kinematic parameter.
A kind of boundling Adjustable calculation method of the low dimensional provided according to the present invention, comprises the following steps：
Step 1：Determine the initial value of kinematic parameter；
Step 2：Minimum calculating, the kinematic parameter after being optimized are carried out to the object function of kinematic parameter；
Step 3：According to the kinematic parameter after optimization, three-dimensional scenic point coordinates is calculated.
Preferably, the step 1 comprises the following steps：
Step 1.1：The dual-view constituted for jth width and the width of jth+1 view, j=1,2 ..., n-1, to the dual-view
On public matching characteristic point set { j, j+1 } corresponding to image characteristic point, using direct linear transformation's algorithm, solve jth+1
Relative pose (R of the width view relative to jth width viewj,j+1,tj,j+1)；
Wherein：
The number of views that n adjusts for participation boundling；
Rj,j+1Relative attitude for the width of jth+1 view relative to jth width view；
tj,j+1Unit relative displacement vector for the width of jth+1 view relative to jth width view, i.e., | | tj,j+1| |=1；
I-th of matching picture point corresponding to public matching characteristic point set { j, j+1 } is calculated in jth width view coordinate system
Under three-dimensional coordinateAnd i-th of matching picture point corresponding to public matching characteristic point set { j, j+1 } is in jth+1
Three-dimensional coordinate under width view coordinate system
Wherein：
I=1,2 ..., m(j,j+1)；
m(j,j+1)Represent the matching picture point in the dual-view of jth width and the width of jth+1 view composition to number；
It is i-th corresponding to public matching characteristic point set { j, j+1 } matching picture point on jth width view
Normalized image point coordinates；
It is i-th corresponding to public matching characteristic point set { j, j+1 } matching picture point on the width view of jth+1
Normalized image point coordinates；
Represent i-th of matching picture point corresponding to public matching characteristic point set { j, j+1 } in jth width view
Three-dimensional coordinate under coordinate system；
Represent i-th of matching picture point corresponding to public matching characteristic point set { j, j+1 } to being regarded in the width of jth+1
Three-dimensional coordinate under figure coordinate system；
Step 1.2：It is fixed | | T1,2| |=1；The three-view diagram constituted for the width of jth -1, jth width and the width of jth+1 view, j
=2,3 ..., n-1, according to the public matching characteristic point set { j-1, j, j+1 } on the three-view diagram, calculates the yardstick of relative displacement |
|Tj,j+1||/||Tj-1,j| |, obtain the unified relative displacement vector T of yardstickj,j+1：
Tj,j+1=| | Tj,j+1||tj,j+1；
Wherein：
T1,2Relative displacement vector for the 2nd width view relative to the 1st width view；
Tj,j+1Relative displacement vector for the width of jth+1 view relative to jth width view；
Tj-1,jRelative displacement vector for jth width view relative to the width view of jth -1；
m(j-1,j,j+1)Represent the public matching image in the three-view diagram that the width of jth -1, jth width and the view of jth+1 width are constituted
Point is to number；
Represent i-th corresponding to the public matching characteristic point set { j-1, j } on the width of jth -1 and jth width view
With picture point to the three-dimensional coordinate under jth width view coordinate system；
Represent i-th corresponding to the public matching characteristic point set { j, j+1 } on jth width and the width view of jth+1
With picture point to the three-dimensional coordinate under jth width view coordinate system；
tj,j+1Unit relative displacement vector for the width of jth+1 view relative to jth width view；
Step 1.3：According to the absolute pose (R of jth width viewj,Tj), calculate the absolute pose for obtaining the width view of jth+1
(Rj+1,Tj+1)：
Rj+1=Rj,j+1Rj
Tj+1=Tj,j+1+Rj,j+1Tj
Wherein：
RjRepresent the absolute pose of jth width view；
Rj+1Represent the absolute pose of the width view of jth+1；
Rj,j+1Relative attitude for the width of jth+1 view relative to jth width view；
TjRepresent the absolute displacement vector of jth width view；
Tj+1Represent the absolute displacement vector of the width view of jth+1；
Tj,j+1Relative displacement vector for the width of jth+1 view relative to jth width view；
When using the first width view as when referring to：
(R1,t1)≡(I3,03×1)
Wherein：
R1Represent the absolute pose of the first width view；
T1Represent the absolute displacement vector of the first width view；
I3Represent the unit matrix of 3-dimensional；
03×1Represent the null matrix of 3 rows 1 row.
Preferably, in the step 2, the object function of the kinematic parameter is specific as follows：
Kinematic parameter θ=(Rj,Tj)J=1,2 ... nMinimum object function δ (θ) be given below：
e3=[0 0 1]T
Wherein：
θ represents the absolute pose parameter set of all views；
δ () represents to minimize object function；
m(j,k)Represent the matching picture point in the dual-view of jth width and kth width view composition to number；
For i-th of matching picture point corresponding to the public matching characteristic point set { j, k } on jth width and kth width view
To the normalized image point coordinates on kth width view；
For i-th of matching picture point corresponding to the public matching characteristic point set { j, k } on jth width and kth width view
To the normalized image point coordinates on jth width view；
Rj,kRelative attitude for kth width view relative to jth width view；
Tj,kRelative displacement vector for kth width view relative to jth width view.
Preferably, the kinematic parameter θ=(R provided in the step 2j,Tj)J=1,2 ... nMinimum object function δ (θ)
Premise be：Identical three dimensional field sight spot is equal to the distance of identical view.
Preferably, the step 3 comprises the following steps：
The kinematic parameter θ=(R obtained according to optimizationj,Tj)J=1,2 ... n, the double vision constituted for jth width and kth width view
Figure, the coordinate at weighted calculation three dimensional field sight spot is as follows：
Tj,k=Tk-Rj,kTj
Wherein：
XiThe three-dimensional coordinate at i-th of three dimensional field sight spot is represented, three dimensional field sight spot XiCorrespondence jth width and kth width view are constituted
Dual-view in s-th of image characteristic point；
Represent i-th of three dimensional field sight spot XiIn the dual-view that jth width and kth width view are constituted whether visible mark
Know function, that is, work as XiWhen visible in the dual-view,Otherwise, then
RjRepresent the absolute pose of jth width view；
Tj,kRelative displacement vector for kth width view relative to jth width view；
Represent s-th of matching picture point corresponding to public matching characteristic point set { j, k } on jth width view
Normalized image point coordinates；
Represent s-th of matching picture point corresponding to public matching characteristic point set { j, k } on kth width view
Normalized image point coordinates；
RkRepresent the absolute pose of kth width view；
TjRepresent the absolute displacement vector of jth width view；
TkRepresent the absolute displacement vector of kth width view；
Rj,kRepresent relative attitude of the kth width view relative to jth width view；
Tj,kRepresent relative displacement vector of the kth width view relative to jth width view.
Preferably, the boundling Adjustable calculation method of the low dimensional, it is considered to the situation that camera has been demarcated, and assume true
The matching picture point pair between each view is determined.
A kind of boundling Adjustable calculation system of the low dimensional provided according to the present invention, includes the meter for the computer program that is stored with
Calculation machine readable storage medium storing program for executing, the computer program realizes the boundling Adjustable calculation side of above-mentioned low dimensional when being executed by processor
The step of method.
Compared with prior art, the present invention has following beneficial effect：
The present invention be it is a kind of initialize that easy, Shandong nation property is good, calculating speed faster, the higher low dimensional boundling of computational accuracy
Method of adjustment.The present invention can be used as the core of the applications such as unmanned vehicle/unmanned plane vision guided navigation, 3 D visual reconstruction, augmented reality
Computing engines.
Brief description of the drawings
By reading the detailed description made with reference to the following drawings to non-limiting example, further feature of the invention,
Objects and advantages will become more apparent upon：
Fig. 1 is the step flow chart of the low dimensional collection beam adjusting method provided according to the present invention.
Embodiment
With reference to specific embodiment, the present invention is described in detail.Following examples will be helpful to the technology of this area
Personnel further understand the present invention, but the invention is not limited in any way.It should be pointed out that to the ordinary skill of this area
For personnel, without departing from the inventive concept of the premise, some changes and improvements can also be made.These belong to the present invention
Protection domain.
The depth of field is expressed as the function of kinematic parameter by the present invention, so that the parameter that three-dimensional scenic point coordinates is adjusted from boundling
Rejected in optimization process.For there is the situation at m three dimensional field sight spot and n width views, parameter space is tieed up for 6*n.Compared to current
Main stream approach, collection beam adjusting method proposed by the present invention significantly reduces the dimension of parameter space.
The present invention considers the situation that camera has been demarcated, and assumes to have determined that the matching picture point pair between each view.
The general type for formula is explained below definition：
It is assumed that n is the number of views that boundling is adjusted, number consecutively is view 1, view 2 ... view n；
(Ri,Ti) represent the i-th width view absolute pose；
RiRepresent the absolute pose of the i-th width view；
Ti=| | Ti||tiRepresent the absolute displacement vector of the i-th width view；
tiRepresent the unit absolute displacement vector of the i-th width view, i.e., | | ti| |=1；
θ represents the absolute pose parameter set of all views；
Represent relative attitude of the kth width view relative to jth width view；
Tj,k≡Tk-RjkTjRepresent relative displacement vector of the kth width view relative to jth width view；
Tj,k=| | Tj,k||tj,k, tj,kUnit relative displacement vector for kth width view relative to jth width view, i.e., | |
tj,k| |=1；
(Rj,k,tj,k) represent relative pose of the kth width view relative to jth width view；
{ j } represents feature point set all on jth width view；
{ j, k } represents the public matching characteristic point set on jth width and kth width view, j, k ... } by that analogy, and represent
Three width are with the public matching characteristic point set on top view；
(j, k) represents the dual-view of jth width and kth width view composition；
m(j,k)Represent the matching picture point in the dual-view of jth width and kth width view composition to number；
I-th of matching picture point respectively in the dual-view of jth width and kth width view composition is in jth
Normalized image point coordinates on width view, kth width view, i.e. the first two component are calibrated image point coordinates, the 3rd
Component is 1.
A kind of low dimensional collection beam adjusting method provided according to the present invention, comprises the following steps：
Step 1：Determine the initial value of kinematic parameter；
Step 2：Minimum calculating, the kinematic parameter after being optimized are carried out to the object function of kinematic parameter；
Step 3：According to the kinematic parameter after optimization, three-dimensional scenic point coordinates is calculated.
Each step is described in detail below.
The step 1 comprises the following steps：
Step 1.1：The dual-view constituted for jth width and the width of jth+1 view, j=1,2 ..., n-1, to the dual-view
On public matching characteristic point set { j, j+1 } corresponding to image characteristic point, using direct linear transformation (DLT, Direct
Linear Transformation) algorithm, solve relative pose (R of the width of jth+1 view relative to jth width viewj,j+1,
tj,j+1)；
Wherein：
The number of views that n adjusts for participation boundling；
Rj,j+1Relative attitude for the width of jth+1 view relative to jth width view；
tj,j+1Unit relative displacement vector for the width of jth+1 view relative to jth width view, i.e., | | tj,j+1| |=1；
I-th of matching picture point corresponding to public matching characteristic point set { j, j+1 } is calculated in jth width view coordinate system
Under three-dimensional coordinateAnd i-th of matching picture point corresponding to public matching characteristic point set { j, j+1 } is in jth+1
Three-dimensional coordinate under width view coordinate system
Wherein：
I=1,2 ..., m(j,j+1)；
m(j,j+1)Represent the matching picture point in the dual-view of jth width and the width of jth+1 view composition to number；
It is i-th corresponding to public matching characteristic point set { j, j+1 } matching picture point on jth width view
Normalized image point coordinates；
It is i-th corresponding to public matching characteristic point set { j, j+1 } matching picture point in the width view of jth+1
On normalized image point coordinates；
Represent i-th of matching picture point corresponding to public matching characteristic point set { j, j+1 } in jth width view
Three-dimensional coordinate under coordinate system；
Represent i-th of matching picture point corresponding to public matching characteristic point set { j, j+1 } to being regarded in the width of jth+1
Three-dimensional coordinate under figure coordinate system；
Step 1.2：Without loss of generality, it is fixed | | T1,2| |=1；For the width of jth -1, jth width and the width view structure of jth+1
Into three-view diagram, j=2,3 ..., n-1 according to the public matching characteristic point set { j-1, j, j+1 } on the three-view diagram, calculate phase
To the yardstick of displacement | | Tj,j+1||/||Tj-1,j| |, obtain the unified relative displacement vector T of yardstickj,j+1：
Tj,j+1=| | Tj,j+1||tj,j+1；
Wherein：
T1,2Relative displacement vector for the 2nd width view relative to the 1st width view；
Tj,j+1Relative displacement vector for the width of jth+1 view relative to jth width view；
Tj-1,jRelative displacement vector for jth width view relative to the width view of jth -1；
m(j-1,j,j+1)Represent the public matching image in the three-view diagram that the width of jth -1, jth width and the view of jth+1 width are constituted
Point is to number；
Represent i-th corresponding to the public matching characteristic point set { j-1, j } on the width of jth -1 and jth width view
With picture point to the three-dimensional coordinate under jth width view coordinate system；
Represent i-th corresponding to the public matching characteristic point set { j, j+1 } on jth width and the width view of jth+1
With picture point to the three-dimensional coordinate under jth width view coordinate system；
tj,j+1Unit relative displacement vector for the width of jth+1 view relative to jth width view；
Step 1.3：According to the absolute pose (R of jth width viewj,Tj), calculate the absolute pose for obtaining the width view of jth+1
(Rj+1,Tj+1)：
Rj+1=Rj,j+1Rj
Tj+1=Tj,j+1+Rj,j+1Tj
Wherein：
RjRepresent the absolute pose of jth width view；
Rj+1Represent the absolute pose of the width view of jth+1；
Rj,j+1Relative attitude for the width of jth+1 view relative to jth width view；
TjRepresent the absolute displacement vector of jth width view；
Tj+1Represent the absolute displacement vector of the width view of jth+1；
Tj,j+1Relative displacement vector for the width of jth+1 view relative to jth width view；
When using the first width view as when referring to：
(R1,t1)≡(I3,03×1)
Wherein：
R1Represent the absolute pose of the first width view；
T1Represent the absolute displacement vector of the first width view；
I3Represent the unit matrix of 3-dimensional；
03×1Represent the null matrix of 3 rows 1 row；
It should be noted that：
-- in step 1.1, j value is j=1,2 ..., n-1；
-- in step 1.2, j value is j=2,3 ..., n-1；
-- in step 1.3, j value is j=1,2 ..., n-1.
In the step 2, the object function of the kinematic parameter is specific as follows：
Identical three dimensional field sight spot to identical view it is equidistant under the premise of, kinematic parameter θ=(Rj,Tj)J=1,2 ... n
Minimum object function δ (θ) be given below：
e3=[0 0 1]T
Wherein：
θ represents the absolute pose parameter set of all views；
δ () represents to minimize object function；
m(j,k)Represent the matching picture point in the dual-view of jth width and kth width view composition to number；
For i-th of matching picture point corresponding to the public matching characteristic point set { j, k } on jth width and kth width view
To the normalized image point coordinates on kth width view；
For i-th of matching picture point corresponding to the public matching characteristic point set { j, k } on jth width and kth width view
To the normalized image point coordinates on jth width view；
Rj,kRelative attitude for kth width view relative to jth width view；
Tj,kRelative displacement vector for kth width view relative to jth width view；
Because the initial value of the kinematic parameter obtained to step 1 by step 2 is optimized, kinematic parameter has been obtained
Optimal value, therefore, step 3 are calculated according to the optimal value of kinematic parameter.Specifically, the step 3 comprises the following steps：
The kinematic parameter θ=(R obtained according to optimizationj,Tj)J=1,2 ... n, the double vision constituted for jth width and kth width view
Figure, the coordinate at weighted calculation three dimensional field sight spot is as follows：
Tj,k=Tk-Rj,kTj
Wherein：
XiThe three-dimensional coordinate at i-th of three dimensional field sight spot is represented, three dimensional field sight spot XiCorrespondence jth width and kth width view are constituted
Dual-view in s-th of image characteristic point；
Represent i-th of three dimensional field sight spot XiIn the dual-view that jth width and kth width view are constituted whether visible mark
Know function, that is, work as XiWhen visible in the dual-view,Otherwise, then
RjRepresent the absolute pose of jth width view；
Tj,kRelative displacement vector for kth width view relative to jth width view；
Represent s-th of matching picture point corresponding to public matching characteristic point set { j, k } on jth width view
Normalized image point coordinates；
Represent s-th of matching picture point corresponding to public matching characteristic point set { j, k } on kth width view
Normalized image point coordinates；
RkRepresent the absolute pose of kth width view；
TjRepresent the absolute displacement vector of jth width view；
TkRepresent the absolute displacement vector of kth width view；
Rj,kRepresent relative attitude of the kth width view relative to jth width view；
Tj,kRepresent relative displacement vector of the kth width view relative to jth width view.
The specific embodiment of the present invention is described above.It is to be appreciated that the invention is not limited in above-mentioned
Particular implementation, those skilled in the art can make a variety of changes or change within the scope of the claims, this not shadow
Ring the substantive content of the present invention.In the case where not conflicting, feature in embodiments herein and embodiment can any phase
Mutually combination.
Claims (7)
1. a kind of boundling Adjustable calculation method of low dimensional, it is characterised in that comprise the following steps：
Step 1：Determine the initial value of kinematic parameter；
Step 2：Minimum calculating, the kinematic parameter after being optimized are carried out to the object function of kinematic parameter；
Step 3：According to the kinematic parameter after optimization, three-dimensional scenic point coordinates is calculated.
2. the boundling Adjustable calculation method of low dimensional according to claim 1, it is characterised in that the step 1 is included such as
Lower step：
Step 1.1：The dual-view constituted for jth width and the width of jth+1 view, j=1,2 ..., n-1, in the dual-view
Image characteristic point corresponding to public matching characteristic point set { j, j+1 }, using direct linear transformation's algorithm, solves the width of jth+1 and regards
Relative pose (R of the figure relative to jth width viewj,j+1,tj,j+1)；
Wherein：
The number of views that n adjusts for participation boundling；
Rj,j+1Relative attitude for the width of jth+1 view relative to jth width view；
tj,j+1Unit relative displacement vector for the width of jth+1 view relative to jth width view, i.e., | | tj,j+1| |=1；
I-th of matching picture point corresponding to public matching characteristic point set { j, j+1 } is calculated under jth width view coordinate system
Three-dimensional coordinateAnd i-th of matching picture point corresponding to public matching characteristic point set { j, j+1 } in the width of jth+1 to regarding
Three-dimensional coordinate under figure coordinate system
<mrow>
<msubsup>
<mi>X</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</msubsup>
<mo>=</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</msubsup>
<mfrac>
<mrow>
<mo>|</mo>
<mo>|</mo>
<msub>
<mi>t</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>&times;</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<mo>|</mo>
</mrow>
<mrow>
<mo>|</mo>
<mo>|</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</msubsup>
<mo>&times;</mo>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<mo>|</mo>
</mrow>
</mfrac>
</mrow>
<mrow>
<msubsup>
<mi>X</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>=</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mfrac>
<mrow>
<mo>|</mo>
<mo>|</mo>
<msub>
<mi>t</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>&times;</mo>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<mo>|</mo>
</mrow>
<mrow>
<mo>|</mo>
<mo>|</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>&times;</mo>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<mo>|</mo>
</mrow>
</mfrac>
</mrow>
Wherein：
I=1,2 ..., m(j,j+1)；
m(j,j+1)Represent the matching picture point in the dual-view of jth width and the width of jth+1 view composition to number；
It is i-th corresponding to public matching characteristic point set { j, j+1 } matching picture point to the normalizing on jth width view
Change image point coordinates；
It is i-th corresponding to public matching characteristic point set { j, j+1 } matching picture point to returning on the width view of jth+1
One changes image point coordinates；
Represent i-th of matching picture point corresponding to public matching characteristic point set { j, j+1 } in jth width view coordinate system
Under three-dimensional coordinate；
Represent i-th of matching picture point corresponding to public matching characteristic point set { j, j+1 } in the width view coordinate of jth+1
Three-dimensional coordinate under system；
Step 1.2：It is fixed | | T1,2| |=1；The three-view diagram constituted for the width of jth -1, jth width and the width of jth+1 view, j=2,
3 ..., n-1, according to the public matching characteristic point set { j-1, j, j+1 } on the three-view diagram, calculates the yardstick of relative displacement | |
Tj,j+1||/||Tj-1,j| |, obtain the unified relative displacement vector T of yardstickj,j+1：
<mrow>
<mo>|</mo>
<mo>|</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>|</mo>
<mo>|</mo>
<mo>/</mo>
<mo>|</mo>
<mo>|</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>j</mi>
<mo>-</mo>
<mn>1</mn>
<mo>,</mo>
<mi>j</mi>
</mrow>
</msub>
<mo>|</mo>
<mo>|</mo>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<msup>
<mi>m</mi>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>-</mo>
<mn>1</mn>
<mo>,</mo>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
</mrow>
</msup>
</munderover>
<msubsup>
<mi>X</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>-</mo>
<mn>1</mn>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>/</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<msup>
<mi>m</mi>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>-</mo>
<mn>1</mn>
<mo>,</mo>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
</mrow>
</msup>
</munderover>
<msubsup>
<mi>X</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>;</mo>
</mrow>
1
Tj,j+1=| | Tj,j+1||tj,j+1；
Wherein：
T1,2Relative displacement vector for the 2nd width view relative to the 1st width view；
Tj,j+1Relative displacement vector for the width of jth+1 view relative to jth width view；
Tj-1,jRelative displacement vector for jth width view relative to the width view of jth -1；
m(j-1,j,j+1)Represent the public matching picture point pair in the three-view diagram that the width of jth -1, jth width and the view of jth+1 width are constituted
Number；
Represent i-th of matching figure corresponding to the public matching characteristic point set { j-1, j } on the width of jth -1 and jth width view
Picture point is to the three-dimensional coordinate under jth width view coordinate system；
Represent i-th of matching figure corresponding to the public matching characteristic point set { j, j+1 } on jth width and the width view of jth+1
Picture point is to the three-dimensional coordinate under jth width view coordinate system；
tj,j+1Unit relative displacement vector for the width of jth+1 view relative to jth width view；
Step 1.3：According to the absolute pose (R of jth width viewj,Tj), calculate the absolute pose (R for obtaining the width view of jth+1j+1,
Tj+1)：
Rj+1=Rj,j+1Rj
Tj+1=Tj,j+1+Rj,j+1Tj
Wherein：
RjRepresent the absolute pose of jth width view；
Rj+1Represent the absolute pose of the width view of jth+1；
Rj,j+1Relative attitude for the width of jth+1 view relative to jth width view；
TjRepresent the absolute displacement vector of jth width view；
Tj+1Represent the absolute displacement vector of the width view of jth+1；
Tj,j+1Relative displacement vector for the width of jth+1 view relative to jth width view；
When using the first width view as when referring to：
(R1,t1)≡(I3,03×1)
Wherein：
R1Represent the absolute pose of the first width view；
T1Represent the absolute displacement vector of the first width view；
I3Represent the unit matrix of 3-dimensional；
03×1Represent the null matrix of 3 rows 1 row.
3. the boundling Adjustable calculation method of low dimensional according to claim 1, it is characterised in that in the step 2, institute
The object function for stating kinematic parameter is specific as follows：
Kinematic parameter θ=(Rj,Tj)J=1,2 ... nMinimum object function δ (θ) be given below：
<mrow>
<mi>&delta;</mi>
<mrow>
<mo>(</mo>
<mi>&theta;</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>n</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</munderover>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<msup>
<mi>m</mi>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msup>
</munderover>
<mrow>
<mo>(</mo>
<mrow>
<mo>|</mo>
<mo>|</mo>
<mfrac>
<mrow>
<msub>
<mi>F</mi>
<mi>L</mi>
</msub>
<mrow>
<mo>(</mo>
<mrow>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>,</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>,</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>,</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
</mrow>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msubsup>
<mi>e</mi>
<mn>3</mn>
<mi>T</mi>
</msubsup>
<mo>&CenterDot;</mo>
<msub>
<mi>F</mi>
<mi>L</mi>
</msub>
<mrow>
<mo>(</mo>
<mrow>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>,</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>,</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>,</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
</mrow>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>-</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
<mo>+</mo>
<mo>|</mo>
<mo>|</mo>
<mfrac>
<mrow>
<msub>
<mi>F</mi>
<mi>R</mi>
</msub>
<mrow>
<mo>(</mo>
<mrow>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>,</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>,</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>,</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
</mrow>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msubsup>
<mi>e</mi>
<mn>3</mn>
<mi>T</mi>
</msubsup>
<mo>&CenterDot;</mo>
<msub>
<mi>F</mi>
<mi>R</mi>
</msub>
<mrow>
<mo>(</mo>
<mrow>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>,</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>,</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>,</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
</mrow>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>-</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
</mrow>
<mo>)</mo>
</mrow>
</mrow>
e3=[0 0 1]T
<mrow>
<msub>
<mi>F</mi>
<mi>L</mi>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>,</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>,</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>,</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mrow>
<mo>|</mo>
<mo>|</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>&times;</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<mo>|</mo>
</mrow>
<mrow>
<mo>|</mo>
<mo>|</mo>
<msubsup>
<mi>X</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>&times;</mo>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<mo>|</mo>
</mrow>
</mfrac>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>+</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
</mrow>
<mrow>
<msub>
<mi>F</mi>
<mi>R</mi>
</msub>
<mrow>
<mo>(</mo>
<mrow>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>,</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>,</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>,</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
</mrow>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mrow>
<mo>|</mo>
<mo>|</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>&times;</mo>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<mo>|</mo>
</mrow>
<mrow>
<mo>|</mo>
<mo>|</mo>
<msubsup>
<mi>X</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>&times;</mo>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<mo>|</mo>
</mrow>
</mfrac>
<msubsup>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mi>T</mi>
</msubsup>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mrow>
<mo>(</mo>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mo>)</mo>
</mrow>
</msubsup>
<mo>-</mo>
<msubsup>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mi>T</mi>
</msubsup>
<msub>
<mi>T</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
</mrow>
Wherein：
θ represents the absolute pose parameter set of all views；
δ () represents to minimize object function；
m(j,k)Represent the matching picture point in the dual-view of jth width and kth width view composition to number；
For i-th corresponding to the public matching characteristic point set { j, k } on jth width and kth width view matching picture point to
Normalized image point coordinates on kth width view；
For i-th corresponding to the public matching characteristic point set { j, k } on jth width and kth width view matching picture point to
Normalized image point coordinates on jth width view；
Rj,kRelative attitude for kth width view relative to jth width view；
Tj,kRelative displacement vector for kth width view relative to jth width view.
4. the boundling Adjustable calculation method of low dimensional according to claim 3, it is characterised in that provided in the step 2
Kinematic parameter θ=(Rj,Tj)J=1,2 ... nMinimum object function δ (θ) premise be：Identical three dimensional field sight spot is regarded to identical
The distance of figure is equal.
5. the boundling Adjustable calculation method of low dimensional according to claim 1, it is characterised in that the step 3 is included such as
Lower step：
The kinematic parameter θ=(R obtained according to optimizationj,Tj)J=1,2 ... n, the dual-view constituted for jth width and kth width view,
The coordinate at weighted calculation three dimensional field sight spot is as follows：
<mrow>
<msub>
<mi>X</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>n</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</munderover>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msubsup>
<mi>&lambda;</mi>
<mi>s</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<msubsup>
<mi>w</mi>
<mi>s</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<msubsup>
<mi>R</mi>
<mi>j</mi>
<mi>T</mi>
</msubsup>
<mrow>
<mo>(</mo>
<mfrac>
<mrow>
<mo>|</mo>
<mo>|</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>&times;</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>s</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<mo>|</mo>
</mrow>
<mrow>
<mo>|</mo>
<mo>|</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>s</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>&times;</mo>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<msubsup>
<mi>x</mi>
<mrow>
<mi>s</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<mo>|</mo>
</mrow>
</mfrac>
<msubsup>
<mi>x</mi>
<mrow>
<mi>s</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>-</mo>
<msub>
<mi>T</mi>
<mi>j</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>=</mo>
<msub>
<mi>R</mi>
<mi>k</mi>
</msub>
<msubsup>
<mi>R</mi>
<mi>j</mi>
<mi>T</mi>
</msubsup>
</mrow>
Tj,k=Tk-Rj,kTj
<mrow>
<msubsup>
<mi>w</mi>
<mi>s</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>=</mo>
<mo>|</mo>
<mo>|</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>s</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>&times;</mo>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<msubsup>
<mi>x</mi>
<mrow>
<mi>s</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
<mo>/</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>n</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</munderover>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msubsup>
<mi>&lambda;</mi>
<mi>s</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<mo>|</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>s</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>&times;</mo>
<msub>
<mi>R</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
</msub>
<msubsup>
<mi>x</mi>
<mrow>
<mi>s</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>,</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
</mrow>
Wherein：
XiThe three-dimensional coordinate at i-th of three dimensional field sight spot is represented, three dimensional field sight spot XiIt is double that correspondence jth width and kth width view are constituted
S-th of image characteristic point in view；
Represent i-th of three dimensional field sight spot XiIn the dual-view that jth width and kth width view are constituted whether visible mark letter
Number, that is, work as XiWhen visible in the dual-view,Otherwise, then
RjRepresent the absolute pose of jth width view；
Tj,kRelative displacement vector for kth width view relative to jth width view；
Represent s-th of matching picture point corresponding to public matching characteristic point set { j, k } to the normalizing on jth width view
Change image point coordinates；
Represent s-th of matching picture point corresponding to public matching characteristic point set { j, k } to the normalizing on kth width view
Change image point coordinates；
RkRepresent the absolute pose of kth width view；
TjRepresent the absolute displacement vector of jth width view；
TkRepresent the absolute displacement vector of kth width view；
Rj,kRepresent relative attitude of the kth width view relative to jth width view；
Tj,kRepresent relative displacement vector of the kth width view relative to jth width view.
6. the boundling Adjustable calculation method of low dimensional according to claim 1, it is characterised in that the boundling of the low dimensional
Adjustable calculation method, it is considered to the situation that camera has been demarcated, and assume to have determined that the matching picture point pair between each view.
7. a kind of boundling Adjustable calculation system of low dimensional, includes the computer-readable recording medium for the computer program that is stored with,
Characterized in that, the computer program realizes the low dimensional any one of claim 1 to 6 when being executed by processor
The step of boundling Adjustable calculation method.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710370360.4A CN107330934B (en) | 2017-05-23 | 2017-05-23 | Low-dimensional cluster adjustment calculation method and system |
| PCT/CN2017/087500 WO2018214179A1 (en) | 2017-05-23 | 2017-06-07 | Low-dimensional bundle adjustment calculation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710370360.4A CN107330934B (en) | 2017-05-23 | 2017-05-23 | Low-dimensional cluster adjustment calculation method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107330934A true CN107330934A (en) | 2017-11-07 |
| CN107330934B CN107330934B (en) | 2021-12-07 |
Family
ID=60192859
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710370360.4A Active CN107330934B (en) | 2017-05-23 | 2017-05-23 | Low-dimensional cluster adjustment calculation method and system |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN107330934B (en) |
| WO (1) | WO2018214179A1 (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109584299A (en) * | 2018-11-13 | 2019-04-05 | 深圳前海达闼云端智能科技有限公司 | Positioning method, positioning device, terminal and storage medium |
| CN109799698A (en) * | 2019-01-30 | 2019-05-24 | 上海交通大学 | The optimal PI parameter optimization method of time lag vision servo system and system |
| CN111161355A (en) * | 2019-12-11 | 2020-05-15 | 上海交通大学 | Pure pose resolving method and system for multi-view camera pose and scene |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060008121A1 (en) * | 2001-06-18 | 2006-01-12 | Microsoft Corporation | Incremental motion estimation through local bundle adjustment |
| US20110311104A1 (en) * | 2010-06-17 | 2011-12-22 | Microsoft Corporation | Multi-Stage Linear Structure from Motion |
| CN106157367A (en) * | 2015-03-23 | 2016-11-23 | 联想(北京)有限公司 | Method for reconstructing three-dimensional scene and equipment |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103985154A (en) * | 2014-04-25 | 2014-08-13 | 北京大学 | Three-dimensional model reestablishment method based on global linear method |
| CN104881869A (en) * | 2015-05-15 | 2015-09-02 | 浙江大学 | Real time panorama tracing and splicing method for mobile platform |
| CN106097436B (en) * | 2016-06-12 | 2019-06-25 | 广西大学 | A kind of three-dimensional rebuilding method of large scene object |
| CN106408653B (en) * | 2016-09-06 | 2021-02-02 | 合肥工业大学 | Real-time robust cluster adjustment method for large-scale three-dimensional reconstruction |
-
2017
- 2017-05-23 CN CN201710370360.4A patent/CN107330934B/en active Active
- 2017-06-07 WO PCT/CN2017/087500 patent/WO2018214179A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060008121A1 (en) * | 2001-06-18 | 2006-01-12 | Microsoft Corporation | Incremental motion estimation through local bundle adjustment |
| US20110311104A1 (en) * | 2010-06-17 | 2011-12-22 | Microsoft Corporation | Multi-Stage Linear Structure from Motion |
| CN106157367A (en) * | 2015-03-23 | 2016-11-23 | 联想(北京)有限公司 | Method for reconstructing three-dimensional scene and equipment |
Non-Patent Citations (2)
| Title |
|---|
| 刘浩敏 等: "基于单目视觉的同时定位与地图构建方法综述", 《计算机辅助设计与图形学学报》 * |
| 齐南: "基于图像序列的目标三维重建技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109584299A (en) * | 2018-11-13 | 2019-04-05 | 深圳前海达闼云端智能科技有限公司 | Positioning method, positioning device, terminal and storage medium |
| CN109584299B (en) * | 2018-11-13 | 2021-01-05 | 深圳前海达闼云端智能科技有限公司 | Positioning method, positioning device, terminal and storage medium |
| CN109799698A (en) * | 2019-01-30 | 2019-05-24 | 上海交通大学 | The optimal PI parameter optimization method of time lag vision servo system and system |
| CN111161355A (en) * | 2019-12-11 | 2020-05-15 | 上海交通大学 | Pure pose resolving method and system for multi-view camera pose and scene |
| CN111161355B (en) * | 2019-12-11 | 2023-05-09 | 上海交通大学 | Multi-view camera pose and scene pure pose resolving method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107330934B (en) | 2021-12-07 |
| WO2018214179A1 (en) | 2018-11-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107564061A (en) | A kind of binocular vision speedometer based on image gradient combined optimization calculates method | |
| CN102426019B (en) | Unmanned aerial vehicle scene matching auxiliary navigation method and system | |
| CN102435188B (en) | Monocular vision/inertia autonomous navigation method for indoor environment | |
| CN110223348A (en) | Robot scene adaptive bit orientation estimation method based on RGB-D camera | |
| CN108648215B (en) | SLAM motion blur pose tracking algorithm based on IMU | |
| CN105976353A (en) | Spatial non-cooperative target pose estimation method based on model and point cloud global matching | |
| CN110008851A (en) | A kind of method and apparatus of lane detection | |
| CN106826833A (en) | Independent navigation robot system based on 3D solid cognition technologies | |
| CN107871327A (en) | The monocular camera pose estimation of feature based dotted line and optimization method and system | |
| CN108534782A (en) | A kind of instant localization method of terrestrial reference map vehicle based on binocular vision system | |
| CN107192375B (en) | A kind of unmanned plane multiple image adaptive location bearing calibration based on posture of taking photo by plane | |
| CN105913410A (en) | Long-distance moving object height measurement apparatus and method based on machine vision | |
| CN106643545A (en) | Calibration method for steel rail profile measured by adopting laser displacement technology | |
| CN103559711A (en) | Motion estimation method based on image features and three-dimensional information of three-dimensional visual system | |
| CN103411589B (en) | A kind of 3-D view matching navigation method based on four-dimensional real number matrix | |
| CN103335648B (en) | A kind of autonomous method for recognising star map | |
| CN106295512A (en) | Many correction line indoor vision data base construction method based on mark and indoor orientation method | |
| CN107063190A (en) | Towards the high-precision direct method estimating of pose of calibration area array cameras image | |
| CN107330934A (en) | The boundling Adjustable calculation method and system of low dimensional | |
| CN111998862A (en) | Dense binocular SLAM method based on BNN | |
| CN115272596A (en) | Multi-sensor fusion SLAM method oriented to monotonous texture-free large scene | |
| CN109724586A (en) | A kind of spacecraft relative pose measurement method of fusion depth map and point cloud | |
| CN108492282A (en) | Three-dimensional glue spreading based on line-structured light and multitask concatenated convolutional neural network detects | |
| CN103854271B (en) | A kind of planar pickup machine scaling method | |
| CN106931978B (en) | Indoor map generation method for automatically constructing road network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |