CN106595602B - Relative orientation method based on homonymous line feature - Google Patents
Relative orientation method based on homonymous line feature Download PDFInfo
- Publication number
- CN106595602B CN106595602B CN201610939617.9A CN201610939617A CN106595602B CN 106595602 B CN106595602 B CN 106595602B CN 201610939617 A CN201610939617 A CN 201610939617A CN 106595602 B CN106595602 B CN 106595602B
- Authority
- CN
- China
- Prior art keywords
- relative orientation
- line
- space
- feature
- formula
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/04—Interpretation of pictures
- G01C11/06—Interpretation of pictures by comparison of two or more pictures of the same area
- G01C11/08—Interpretation of pictures by comparison of two or more pictures of the same area the pictures not being supported in the same relative position as when they were taken
Abstract
The present invention relates to a kind of relative orientation methods based on homonymous line feature.Firstly, acquiring homonymous line feature in stereogram；By the relative orientation equations simultaneousness of parallel lines feature, vertical line feature and intersecting straight lines feature, linearisation in the case where given initial value, is iterated calculating using the method for least square adjustment, refines and determine the relative displacement between two imagestAnd rotationR.The present invention does not need traditional point feature of the same name, does not introduce any additional parameter, rely solely on the correlation between homonymous line feature in solution process, can resolve relative displacement and rotation between two images；Be able to solve the poor region of texture due to same place is insufficient and the problem of can not accurately orient, can be applied to cybercity construction, industrial photogrammetry, Lidar point cloud data and be registrated with aviation image etc..
Description
Technical field
The present invention relates to a kind of relative orientation calculation methods, more particularly, to the relative orientation solution based on homonymous line feature
Calculation method.
Background technique
Relative orientation be restore stereogram between relative displacement and posture a kind of method, be restore core line geometry and
The important means of camera geometry is photogrammetric and the unfailing research hotspot of computer vision field.It is deposited in stereogram
In line characteristic information abundant.Line feature can also provide mutual between line and line other than it can describe the profile of object
Location information.The poor region of texture on image, is difficult to find out reliable point feature of the same name, but the line feature of the same area
It is often more obvious.Therefore, compared with traditional point feature, line feature can solve the poor region same place of texture better
It is rare and the problem of can not accurately orient.In view of above-mentioned advantage, the relative orientation based on line feature has caused photogrammetric neck
The concern in domain.
Currently, the relative orientation based on line feature has had certain development, parallelogram, broad sense can use
Line feature is added relative orientation model, carries out least square adjustment resolving by the methods of Hough transform, polar coordinates, Generalized Point.
Above method using the geometrical characteristic of line as constraint, is transformed conditional equation.However, these methods still need it is of the same name
The auxiliary of point, the orientation for not being suitable for the poor region of texture resolves, and introduces more additional parameter in calculating process, causes
The unknown excessive, calculation amount that keeps count of increases, resolves the problems such as difficult.Meanwhile there is sternly the selection of linear feature in above method
The limitation of lattice, linear ground object need to meet specific geometrical constraint, cause the not versatility of method.
Linear feature is the most common line feature, is common in house, road, farmland, industrial part etc..Straight line and straight line
Relationship can be generally divided into three kinds: parallel, vertical and intersection.Therefore, if parallel between capable of making full use of straight line, vertical and intersection is closed
System, constructs new relative orientation model, it should can be completely free of the constraint of point feature of the same name, enhance the versatility of algorithm, be
The application services such as camera geometric orientation.
Summary of the invention
The present invention is mainly the geometric orientation solved the problems, such as between camera, proposes a kind of phase based on homonymous line feature
To orientation method, can be constructed according to parallel lines feature of the same name, vertical line feature of the same name and intersecting straight lines feature of the same name
New relative orientation model resolves the relative orientation parameter between camera, including the displacement t and rotation R between camera.Spin moment
Battle array R can usually use Eulerian anglesω, κ are indicated；(μ, ν, 1) usually can be represented by vectors in displacement tTIt indicates.Therefore, usually will
Elements of relative orientation is divided into angle elementω, κ and line element μ, ν.The purpose of relative orientation is smart between camera in order to resolve
True elements of relative orientation.
Above-mentioned technical problem of the invention is mainly to be addressed by following technical proposals:
A kind of relative orientation method based on homonymous line feature, which is characterized in that by linear ground object in stereogram
Gray feature be defined as homonymous line feature, specifically:
Enable L1, L2Two object space straight lines are indicated, by object space straight line L1Back projection divides on each image to stereogram
Not Dui Ying an image space straight line；This two image space straight lines are defined as l1And l1'；Similarly, by object space straight line L2Back projection to solid
As right, two image space straight lines are also corresponded to, l is respectively defined as2And l2'；Due to image space straight line l1And l1' corresponding same object space is straight
Line L1, therefore straight line l1And l1' it is homonymous line；Similarly, l2And l2' it is also homonymous line；Object space straight line L1, L2In space may be used
Can there are different geometric topo-relationships, including parallel relation, vertical relation and overlapping relation, each relationship can be listed
One relative orientation equation resolves service for camera geometric orientation；
The relative orientation method the following steps are included:
Step 1, the stereogram of image composition is opened according to the left and right of input two, extracts linear feature, including of the same name parallel
Linear feature, vertical line feature of the same name and intersecting straight lines feature of the same name；
Step 2, according to different linear features, different relative orientation models is listed respectively；
Step 3, all relative orientation models of simultaneous, linearisation construct error equation group；
Step 4, using the method for least square, the relative orientation parameter between camera is iterated to calculate.
In a kind of above-mentioned relative orientation method based on homonymous line feature, the step 2, according to L1, L2It is flat
Capable or vertical or intersection, defines relative orientation model respectively；
Situation one: L1And L2It is parallel: to work as L1And L2In parallel, there are relational expressions
In formula, R indicates the spin matrix of relative orientation, for expressing the rotation between camera；Indicate intrinsic parameter square
Image space rectilinear coordinates under battle array is unitization；
By formula (1) as it can be seen that available two equations of a pair of of parallel lines；In the case where there is initial value, can change
In generation, calculates spin matrix R；Since space vector is unrelated with displacement, parallel relation is relied solely on, is that can not resolve phase seat in the plane
Move t；
Situation two: L1And L2It is vertical: as space line L1And L2Vertically, available relational expression
In formula, R indicates the spin matrix of relative orientation；Indicate that the image space straight line under Intrinsic Matrix is unitization is sat
Mark；
By above formula as it can be seen that according to vertical relation, an equation is can be obtained in a pair of mutually perpendicular line；There is the case where initial value
Under, it can be iterated and spin matrix R is calculated；Since space vector is unrelated with displacement, vertical relation is relied solely on,
It is that can not resolve camera displacement t；It needs to calculate translation vector t using new rectilinear geometry relationship；
Situation three: L1And L2Intersection, available relational expression:
In formula, symbol is indicated the coordinate components inverted order of vector；L1, L2Indicate the Plucker coordinate of object space straight line；
Enable S1, S2Respectively indicate the photo centre of left and right image camera；Still indicate that image space is straight using 3 × 1 homogeneous vectors
Line l1,l2,l1',l2'；；Enable πiI=1,2 indicates photo centre S1With straight line liThe plane constituted, πi' indicate photo centre S2
With straight line li' the plane that is constituted；π1,π1',π2,π2' can be indicated respectively with one 4 × 1 homogeneous vectors:
Wherein,Indicate the image space rectilinear coordinates under Intrinsic Matrix is unitization；Respectively indicate left and right
Camera matrix of the image in the case where Intrinsic Matrix is unitization；It is assumed that using the image space coordinate system of left image as world coordinate system,
The camera matrix of left and right image respectively indicates are as follows:
R is with relative orientation angle elementThe spin matrix of ω, κ composition；T=(1, μ, ν)TIndicate relative orientation line element
Element；I indicates 3 × 3 unit matrix；0 indicates 3 × 1 null vector；
Space line L1、L2It is plane π respectively1,π1' between and plane π2,π2' between intersection；Space line L1, L2's
Plucker coordinate can indicate are as follows:
Wherein,Indicate straight line L1, L2Plucker coordinate components；Vec () is indicated will
4 × 4 Plucker matrix-vector；
Formula (3), formula (4) and formula (5) indicate that when straight line intersection, identified relative orientation model not only includes rotation
Matrix R also deviates t comprising camera, it is only necessary to which at least two pairs of a intersecting straight lines of the same name can carry out resolving t.
In a kind of above-mentioned relative orientation method based on homonymous line feature, the step 3, error equation group
Simultaneous mode are as follows:
As space line L1With L2When parallel, formula (10) can be arranged are as follows:
As space line L1With L2When vertical, formula (14) can be arranged are as follows:
As space line L1With L2When intersection, formula (17) can be arranged are as follows:
Simultaneous Equations { F1,F2,F3,F4, available one one about relative orientation angle element and line element is non-
System of linear equations；In the presence of initial value, equation group can be linearized, realize that adjustment resolves:
In formula, Fi,0It indicates the initial value of elements of relative orientation substituting into FiThe result obtained afterwards；Δω,Δκ,Δμ,Δ
ν is respectively the correction of elements of relative orientation；
Each conditional equation FiIt is all different to the local derviation of elements of relative orientation, as follows respectively:
In a kind of above-mentioned relative orientation method based on homonymous line feature, the step 4, least square is resolved
Method is as follows:
According to formula (21), (22), (23) and (24), normal equation is formed:
In formula, V indicates residual vector；A representation equation coefficient matrix；L indicates constant vector；Δ X indicates that unknown number changes
Positive quantity；
After calculating unknown number reduction Δ X, unknown number reduction Δ X is added into unknown number initial value X0, can be obtained more
Add accurate elements of relative orientation；The general method for using iteration, constantly refines elements of relative orientation；
X1=X0+ΔX (26)
After elements of relative orientation has been calculated, according to angle elementω, κ can calculate spin matrix R:
Therefore, the present invention has the advantage that being not necessarily to only according to parallel, vertical and intersection homonymous line feature
Same place can determine the elements of relative orientation between stereogram, be able to solve the same places such as poor, the repetition texture of texture
Relative orientation problem with difficult region.The present invention is in space photogrammetry, aerophotogrammetry, low altitude photogrammetry and close shot
There is preferable application prospect in photogrammetric field.
Detailed description of the invention
Fig. 1 is the relative orientation calculation method specific flow chart based on homonymous line feature.
Specific embodiment
Below with reference to the embodiments and with reference to the accompanying drawing the technical solutions of the present invention will be further described.
Embodiment:
Technical solution provided by the invention is, for the situation that linear feature in image is abundant, according between space line
Parallel, vertical or overlapping relation, propose a kind of relative orientation method based on homonymous line feature, can be according between straight line
Geometrical relationship, establish new relative orientation model, resolve the relative orientation parameter between camera, be able to solve due to same place
It is rare and the problem of can not accurately orient.The implementation route of the program is as shown in Figure 1, comprising the following steps:
Step 1. extracts linear feature on image.
There are linear features abundant on image, can manually extract, and can also automatically extract straight line by canny operator
Feature.
Step 2. lists different relative orientation models according to different linear features respectively.
The straight line on image is expressed using the homogeneous coordinates in projective geometry.According to the following formula, the straight line on image is sat
Mark be converted to Intrinsic Matrix it is unitization under rectilinear coordinates.The purpose for the arrangement is that in the case where not influencing calculated result,
So that subsequent computational short cut.
Wherein, K indicates Intrinsic Matrix；liIndicate the straight line of left image；li' indicate the straight line of right image；Respectively indicate the rectilinear coordinates under unit matrix intrinsic parameter.
When straight line parallel, the relative orientation model based on parallel lines is listed according to the following formula:
When straight line is vertical, the relative orientation model based on vertical line is listed according to the following formula:
When straight line intersection, the relative orientation model based on intersecting straight lines is listed according to the following formula:
By all model simultaneous, a Nonlinear System of Equations is formed.
Step 3. linearizes the Nonlinear System of Equations in step 2, forms error equation group.
Elements of relative orientation includes three angle elementsω, κ and two line elements μ, ν.The purpose of relative orientation is to resolve
The accurate solution of this five elements.According to the initial value of elements of relative orientation, according to the following formula, to the Nonlinear System of Equations line in step 2
Property:
In formula,Δ ω, Δ κ, Δ μ, Δ ν are respectively the correction of elements of relative orientation；The coefficient of each correction
It is expressed as follows:
Step 4. uses least square method, calculates the relative orientation parameter between camera.
After Nonlinear System of Equations is linearized, the system of linear equations being shown below is constituted:
V=A Δ X-L
V=(v1 v2 v3 ...)T
In formula, V indicates residual vector；A representation equation coefficient matrix；L indicates constant vector；Δ X indicates that unknown number changes
Positive quantity.
Above-mentioned system of linear equations is solved using the method for least square, is shown below:
Δ X=(ATA)-1ATL
After calculating unknown number reduction Δ X, unknown number reduction Δ X is added to the initial value X of elements of relative orientation0, i.e.,
More accurate elements of relative orientation can be obtained.The general method for using iteration, constantly refines elements of relative orientation, such as following formula
It is shown:
X1=X0+ΔX
After elements of relative orientation has been calculated, according to angle elementω, κ can calculate spin matrix R:
Specific embodiment described herein is only an example for the spirit of the invention.The neck of technology belonging to the present invention
The technical staff in domain can make various modifications or additions to the described embodiments or replace by a similar method
In generation, however, it does not deviate from the spirit of the invention or beyond the scope of the appended claims.
Claims (3)
1. a kind of relative orientation method based on homonymous line feature, which is characterized in that by linear ground object in stereogram
Gray feature is defined as homonymous line feature, specifically:
Enable L1, L2Two object space straight lines are indicated, by object space straight line L1Back projection is right respectively on each image to stereogram
Answer the straight line of an image space；This two image space straight lines are defined as l1And l1'；Similarly, by object space straight line L2Back projection is to space image
It is right, two image space straight lines are also corresponded to, l is respectively defined as2And l2'；Due to image space straight line l1And l1' correspond to same object space straight line
L1, therefore straight line l1And l1' it is homonymous line；Similarly, l2And l2' it is also homonymous line；Object space straight line L1, L2In space may
There are different geometric topo-relationships, including parallel relation, vertical relation and overlapping relation, each relationship can list one
A relative orientation equation resolves service for camera geometric orientation；
The relative orientation method the following steps are included:
Step 1, the stereogram of image composition is opened according to the left and right of input two, extracts linear feature, including parallel lines of the same name
Feature, vertical line feature of the same name and intersecting straight lines feature of the same name；
Step 2, according to different linear features, different relative orientation models is listed respectively；
Step 3, all relative orientation models of simultaneous, linearisation construct error equation group；
Step 4, using the method for least square, the relative orientation parameter between camera is iterated to calculate；
In the step 2, according to L1, L2Parallel or vertical or intersection, define relative orientation model respectively；
Situation one: L1And L2It is parallel: to work as L1And L2In parallel, there are relational expressions
In formula, R indicates the spin matrix of relative orientation, for expressing the rotation between camera；Indicate Intrinsic Matrix list
Image space rectilinear coordinates under positionization；
By formula (1) as it can be seen that available two equations of a pair of of parallel lines；In the case where there is initial value, meter can be iterated
Calculate spin matrix R；Since space vector is unrelated with displacement, parallel relation is relied solely on, is that can not resolve camera offset t；
Situation two: L1And L2It is vertical: as space line L1And L2Vertically, available relational expression
In formula, R indicates the spin matrix of relative orientation；Indicate the image space rectilinear coordinates under Intrinsic Matrix is unitization；
By above formula as it can be seen that according to vertical relation, an equation is can be obtained in a pair of mutually perpendicular line；In the case where there is initial value,
It can be iterated and spin matrix R is calculated；Since space vector is unrelated with displacement, vertical relation is relied solely on, is
Camera offset t can not be resolved；Need to calculate camera offset t using new rectilinear geometry relationship；
Situation three: L1And L2Intersection, available relational expression:
In formula, symbol-expression is by the coordinate components inverted order of vector；L1, L2Indicate the Plucker coordinate of object space straight line；
Enable S1, S2Respectively indicate the photo centre of left and right image camera；Still image space straight line is indicated using 3 × 1 homogeneous vectors
l1,l2,l1',l2'；Enable πiIndicate photo centre S1With straight line liThe plane constituted, πi' indicate photo centre S2With straight line li'
The plane constituted, wherein i=1,2；π1,π1',π2,π2' can be indicated respectively with one 4 × 1 homogeneous vectors:
Wherein,Indicate the image space rectilinear coordinates under Intrinsic Matrix is unitization；Respectively indicate left and right image
Camera matrix in the case where Intrinsic Matrix is unitization；It is assumed that using the image space coordinate system of left image as world coordinate system, left and right
The camera matrix of image respectively indicates are as follows:
R is with relative orientation angle elementThe spin matrix of ω, κ composition；T=(1, μ, ν)ΤIndicate relative orientation line element；I table
Show 3 × 3 unit matrix；0 indicates 3 × 1 null vector；
Space line L1、L2It is plane π respectively1,π1' between and plane π2,π2' between intersection；Space line L1, L2's
Plucker coordinate can indicate are as follows:
Wherein,Indicate straight line L1, L2Plucker coordinate components；Vec () is indicated 4 × 4
Plucker matrix-vector, wherein i=1,2；
Formula (3), formula (4) and formula (5) indicate that when straight line intersection, identified relative orientation model not only includes spin matrix
R also deviates t comprising camera, it is only necessary to which at least two pairs of a intersecting straight lines of the same name can carry out resolving t.
2. a kind of relative orientation method based on homonymous line feature according to claim 1, which is characterized in that described
In step 3, the simultaneous mode of error equation group are as follows:
As space line L1With L2When parallel, formula (1) can be arranged are as follows:
As space line L1With L2When vertical, formula (2) can be arranged are as follows:
As space line L1With L2When intersection, formula (3) can be arranged are as follows:
Simultaneous Equations { F1,F2,F3,F4, available one one about relative orientation angle element and line element is non-linear
Equation group；In the presence of initial value, equation group can be linearized, realize that adjustment resolves:
In formula, Fi,0It indicates the initial value of elements of relative orientation substituting into FiThe result obtained afterwards；Δ ω, Δ κ, Δ μ, Δ ν points
Not Wei elements of relative orientation correction；
Each conditional equation FiIt is all different to the local derviation of elements of relative orientation, as follows respectively:
3. a kind of relative orientation method based on homonymous line feature according to claim 2, which is characterized in that described
In step 4, it is as follows that least square resolves method:
According to formula (21), (22), (23) and (24), normal equation is formed:
In formula, V indicates residual vector；A representation equation coefficient matrix；L indicates constant vector；Δ X indicates unknown number reduction；
After calculating unknown number reduction Δ X, unknown number reduction Δ X is added into unknown number initial value X0, can be obtained more accurate
Elements of relative orientation；The general method for using iteration, constantly refines elements of relative orientation；
X1=X0+ΔX (26)
After elements of relative orientation has been calculated, according to angle elementω, κ can calculate spin matrix R:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610939617.9A CN106595602B (en) | 2016-10-31 | 2016-10-31 | Relative orientation method based on homonymous line feature |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610939617.9A CN106595602B (en) | 2016-10-31 | 2016-10-31 | Relative orientation method based on homonymous line feature |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106595602A CN106595602A (en) | 2017-04-26 |
| CN106595602B true CN106595602B (en) | 2019-06-25 |
Family
ID=58589681
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610939617.9A Active CN106595602B (en) | 2016-10-31 | 2016-10-31 | Relative orientation method based on homonymous line feature |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106595602B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110288636B (en) * | 2019-05-05 | 2020-02-18 | 中国矿业大学 | LiDAR point cloud non-initial value registration method based on plane feature constraint |
| CN110443837B (en) * | 2019-07-03 | 2021-09-24 | 湖北省电力勘测设计院有限公司 | Urban area airborne laser point cloud and aerial image registration method and system under constraint of linear features |
| CN111126450B (en) * | 2019-11-29 | 2024-03-19 | 上海宇航系统工程研究所 | Modeling method and device for cuboid space vehicle based on nine-line configuration |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102147249A (en) * | 2010-02-08 | 2011-08-10 | 同济大学 | Method for precisely correcting satellite-borne optical linear array image based on linear characteristic |
| CN102419172A (en) * | 2011-08-18 | 2012-04-18 | 武汉大学 | Stereo image pair automatic relative orientation method with additional non-linear constraint condition |
| CN102519484A (en) * | 2011-11-29 | 2012-06-27 | 武汉大学 | Multi-disc overall adjustment calibration method of rotary photogrammetry system |
| CN105354853A (en) * | 2015-11-30 | 2016-02-24 | 中国人民解放军国防科学技术大学 | Non-coplanar control straight line imaging based two-step method for calibrating camera parameters |
-
2016
- 2016-10-31 CN CN201610939617.9A patent/CN106595602B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102147249A (en) * | 2010-02-08 | 2011-08-10 | 同济大学 | Method for precisely correcting satellite-borne optical linear array image based on linear characteristic |
| CN102419172A (en) * | 2011-08-18 | 2012-04-18 | 武汉大学 | Stereo image pair automatic relative orientation method with additional non-linear constraint condition |
| CN102519484A (en) * | 2011-11-29 | 2012-06-27 | 武汉大学 | Multi-disc overall adjustment calibration method of rotary photogrammetry system |
| CN105354853A (en) * | 2015-11-30 | 2016-02-24 | 中国人民解放军国防科学技术大学 | Non-coplanar control straight line imaging based two-step method for calibrating camera parameters |
Non-Patent Citations (3)
| Title |
|---|
| "基于多种同名特征的相对定向方法研究";张永军等;《测绘学报》;20110430;第40卷(第2期);第194-199页 |
| "基于相交直线的相对定向方法";张永军等;《武汉大学学报﹒信息科学版》;20150331;第40卷(第3期);正文第303页第1段至第306页最后一段及图1-6 |
| "改进的同名直线匹配及三维重建技术";王建敏等;《海洋测绘》;20110331;第31卷(第2期);第68-71页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106595602A (en) | 2017-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107564061B (en) | Binocular vision mileage calculation method based on image gradient joint optimization | |
| EP3274964B1 (en) | Automatic connection of images using visual features | |
| CN102646275A (en) | Method for realizing virtual three-dimensional superposition through tracking and positioning algorithms | |
| CN111459269B (en) | Augmented reality display method, system and computer readable storage medium | |
| CN106595602B (en) | Relative orientation method based on homonymous line feature | |
| CN104077760A (en) | Rapid splicing system for aerial photogrammetry and implementing method thereof | |
| CN104616247B (en) | A kind of method for map splicing of being taken photo by plane based on super-pixel SIFT | |
| CN105354396A (en) | Geometric comparison method for models obtained through collaborative modelling of different software | |
| WO2019223262A1 (en) | Depth recovery method and apparatus for monocular images and computer device | |
| CN109214314B (en) | Automatic fusion matching algorithm for lane lines | |
| Tao et al. | Stereo priori RCNN based car detection on point level for autonomous driving | |
| CN109974743A (en) | A kind of RGB-D visual odometry optimized based on GMS characteristic matching and sliding window pose figure | |
| CN113920263A (en) | Map construction method, map construction device, map construction equipment and storage medium | |
| CN112733641A (en) | Object size measuring method, device, equipment and storage medium | |
| CN104318552A (en) | Convex hull projection graph matching based model registration method | |
| CN105787464A (en) | A viewpoint calibration method of a large number of pictures in a three-dimensional scene | |
| Feng et al. | Semi-automatic 3d reconstruction of piecewise planar building models from single image | |
| Gokhool et al. | A dense map building approach from spherical RGBD images | |
| Jahanshahi et al. | Accurate and robust scene reconstruction in the presence of misassociated features for aerial sensing | |
| CN111553969B (en) | Texture mapping method, medium, terminal and device based on gradient domain | |
| Lee et al. | Robust uncertainty-aware multiview triangulation | |
| Carceroni et al. | Structure from motion with known camera positions | |
| CN109782755A (en) | Control AGV calibrated, the method for AGV calibrating position, computer storage medium and AGV | |
| CN110570473A (en) | weight self-adaptive posture estimation method based on point-line fusion | |
| Yan et al. | A novel bundle adjustment method with additional ground control point constraint |
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 |