CN110335296A - A kind of point cloud registration method based on hand and eye calibrating - Google Patents
A kind of point cloud registration method based on hand and eye calibrating Download PDFInfo
- Publication number
- CN110335296A CN110335296A CN201910544860.4A CN201910544860A CN110335296A CN 110335296 A CN110335296 A CN 110335296A CN 201910544860 A CN201910544860 A CN 201910544860A CN 110335296 A CN110335296 A CN 110335296A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- point
- coordinate system
- measurement
- measuring device
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention belongs to three-dimensional measurement fields, and specifically disclose a kind of point cloud registration method based on hand and eye calibrating, it drives measuring device to be moved twice by robot end, and respectively obtain the transformation matrix of the transformation matrix and measuring device coordinate system of robot ending coordinates system when moving twice, then according to hand and eye calibrating equation, acquire robot end's coordinate system to measuring device coordinate system transformation matrix, and then obtain robot base coordinate sys-tem to measuring device coordinate system transformation matrix, further according to this transformation matrix, the object under test measurement pointcloud that measuring device obtains is transformed into robot base coordinate sys-tem, complete the point cloud rough registration between measurement pointcloud and model point cloud, point cloud essence registration is completed finally by with the relationship between matching double points in the measurement pointcloud and model point cloud under robot base coordinate sys-tem.Whole process be not required to testee surface paste index point, simplify point cloud registering process, and may be implemented measurement high temp objects when point cloud registering.
Description
Technical field
The invention belongs to three-dimensional measurement fields, more particularly, to a kind of point cloud registration method based on hand and eye calibrating.
Background technique
Point cloud registering is to arrive the point cloud data under different perspectives by rigid transformations unified integrations such as rotation translations
Process under specified coordinate system is mainly used for the three-dimensional reconstruction of testee.Point cloud registering includes rough registration and essence registration: thick
Registration is in the case where source point cloud and target point cloud do not know any initial relative position completely, and quickly estimation one is rough
The method for registering of point cloud registering matrix；Essence registration is to be counted recently using the initial transformation matrix in rough registration by such as iteration
The method for registering more accurately solved is calculated in the methods of method (ICP algorithm).
Existing point cloud registration method needs to paste index point on testee surface, passes through the mark between two consecutive points clouds
Point is but more time-consuming in testee surface patch index point to realize the rough registration of a cloud, is unfavorable for automatized three-dimensional
Measuring system is built, and for high temperature workpiece, and on its surface, patch index point is unpractical.Therefore, for high temperature
For workpiece automatized three-dimensional measuring system, how to accomplish not by label will point cloud can be carried out registration be one urgently
Problem to be solved.
Summary of the invention
Aiming at the above defects or improvement requirements of the prior art, the present invention provides a kind of, and the point cloud based on hand and eye calibrating is matched
Quasi- method is sat it is intended that the measurement pointcloud under measuring device coordinate system is transferred to robot base based on hand and eye calibrating
Under mark system, so that the rough registration of a cloud is realized, then by with the measurement pointcloud and model points under robot base coordinate sys-tem
Relationship between matching double points in cloud completes point cloud essence registration, and whole process does not need to paste index point, letter on testee surface
Change point cloud registering process and result is accurate.
To achieve the above object, it is an aspect of this invention to provide that the invention proposes a kind of point cloud based on hand and eye calibrating
Method for registering includes the following steps:
S1 measuring device is mounted on robot end, and is driven by robot end from initial position and be moved to the first measurement
Point obtains transformation matrix A of robot end's coordinate system from initial position to the first measurement point1And measuring device coordinate system
From initial position to the transformation matrix B of the first measurement point1；
S2 robot end drives measuring device to be moved to the second measurement point from the first measurement point, obtains robot end's seat
Transformation matrix A of the mark system from the first measurement point to the second measurement point2And measuring device coordinate system is from the first measurement point to second
The transformation matrix B of measurement point2；
S3 is according to hand and eye calibrating equation, by the transformation matrix A1、B1、A2、B2Acquire same position robot end's coordinate
It is the transformation matrix to measuring device coordinate system；
S4 by robot end's coordinate system to measuring device coordinate system transformation matrix and known robot base
Seat coordinate system obtains the change of robot base coordinate sys-tem to measuring device coordinate system to the transformation matrix of robot end's coordinate system
Change matrix；Then according to this transformation matrix, the object under test measurement pointcloud that measuring device is obtained turns from measuring device coordinate system
It changes in robot base coordinate sys-tem, completes the point cloud rough registration between measurement pointcloud and the model point cloud of object under test；
S5 is obtained with the matching double points in the measurement pointcloud and model point cloud under robot base coordinate sys-tem, according to matching
Relationship between point pair completes point cloud essence registration.
As it is further preferred that hand and eye calibrating equation in the S3 specifically:
Wherein, X is transformation matrix of robot end's coordinate system to measuring device coordinate system.
As it is further preferred that solving hand and eye calibrating equation using Matrix Direct Product method.
As it is further preferred that the S5 specifically comprises the following steps:
(1) for each point in model point cloud, its nearest neighbor point is searched in measurement pointcloud respectively as match point, from
And multipair matching double points are obtained, form matching double points collection；
(2) average value of distance between all matching double points is calculatedWith standard deviation, when distance between two points in matching double pointsMeetThis matching double points is then retained in matching double points to concentrate, is otherwise then rejected, to update
Matching double points collection；Wherein, s is the error scale factor；
(3) objective function is constructed:
Wherein, piWith qiIt is a pair of of point that matching double points are concentrated, and piFor the point in measurement pointcloud, qiFor in model point cloud
Point, niIt is qiNormal vector, NCIt is the quantity that matching double points concentrate matching double points；To acquire measurement pointcloud to model point cloud
Spin matrix R and translation matrix T, and measurement pointcloud is updated according to this spin matrix R and translation matrix T；
(4) it repeats step (1) and arrives (3), when meeting the preset condition of convergence, stop iteration, obtain final spin moment
Battle array and translation matrix complete point cloud essence registration.
As it is further preferred that retaining distance between two points in matching double points after the step (2) update matching double points collectionLesser part matching double points update matching double points collection again.
As it is further preferred that the condition of convergence are as follows: when iteration, the cosine value of measurement pointcloud rotation angle is greater than
The translation distance of preset threshold rotating value and measurement pointcloud be less than preset translation threshold value when or the number of iterations reach it is preset most
When big the number of iterations, objective function convergence stops iteration.
In general, through the invention it is contemplated above technical scheme is compared with the prior art, mainly have below
Technological merit:
1. being not required to paste on testee surface in this course the present invention is based on the rough registration that hand and eye calibrating realizes point cloud
Index point simplifies point cloud registering process, and may be implemented measurement high temp objects when point cloud registering.
2. the present invention is solved using the hand and eye calibrating based on Matrix Direct Product, eliminates the error propagation generated when decoupling and tire out
Long-pending problem, compared with other hand and eye calibrating methods, this method has the characteristics that precision is high, high-efficient, robustness is good.
3. the present invention is carrying out a cloud essence with punctual using the rotation condition of convergence and the translation condition of convergence, tradition can be excluded
When root-mean-square distance (root-mean-square distance, RMS) condition of convergence that ICP algorithm uses is judged, cloud is put
Registration reality has restrained, but causes RMS value to become larger to be judged as not converged and continue iteration since Independent Point is not rejected completely
The case where.
Detailed description of the invention
Fig. 1 is measurement method and coordinate system schematic diagram in the embodiment of the present invention；
Fig. 2 is midpoint of embodiment of the present invention cloud essence registration flow chart.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.As long as in addition, technical characteristic involved in the various embodiments of the present invention described below
Not constituting a conflict with each other can be combined with each other.
A kind of point cloud registration method based on hand and eye calibrating provided in an embodiment of the present invention, includes the following steps:
S1 measuring device (such as camera) is mounted on robot end, and is driven by robot end from initial position movement
To the first measurement point, transformation matrix A of robot end's coordinate system from initial position to the first measurement point is obtained1, and measurement
Transformation matrix B of the device coordinate system from initial position to the first measurement point1；
S2 robot end drives measuring device to be moved to the second measurement point from the first measurement point, obtains robot end's seat
Transformation matrix A of the mark system from the first measurement point to the second measurement point2And measuring device coordinate system is from the first measurement point to second
The transformation matrix B of measurement point2；
S3 is when robot end drives measuring device mobile, change of robot end's coordinate system to measuring device coordinate system
It changes matrix X not change, and according to hand and eye calibrating equation, obtains:
Specifically, Wherein, RX、RA1、RB1、RA2、RB2It is 3 × 3 unit orthogonal matrix for corresponding spin matrix；
TX、TA1、TB1、TA2、TB2It is 3 × 1 matrix for corresponding translation matrix；
Then formula (1) can convert are as follows:
Formula (2) are solved using Matrix Direct Product method, obtain the transformation of robot end's coordinate system to measuring device coordinate system
Matrix:
X=(ATA)-1ATb (3)
Wherein,09×3
Represent 9 rows 3 column null matrix, 09Represent the null vector of 9 rows 1 column.
The transformation matrix Y of S4 robot base coordinate sys-tem to robot end's coordinate system can be obtained directly from robot,
By robot end's coordinate system in transformation matrix Y and S3 to the transformation matrix X of measuring device coordinate system, robot base is obtained
Transformation matrix Z=YX of the coordinate system to measuring device coordinate system；
The point in object under test measurement pointcloud then obtained for measuring device, the coordinate under measuring device coordinate system
For (xi, yi, zi), the coordinate under robot base coordinate sys-tem is (x 'i, y 'i, z 'i), have:
The object under test measurement pointcloud that measuring device obtains can be transformed into robot base from measuring device coordinate system
In coordinate system, (model point cloud is to first pass through CAD in advance to the point cloud rough registration between completion measurement pointcloud and the model point cloud of object under test
Etc. data of the object under test that obtains of modes under base machine people's base coordinate system)；
Specifically, as shown in Figure 1, ObXbYbZbFor robot base coordinate sys-tem, OtXtYtZtFor robot end's coordinate system,
OsXsYsZsFor measuring device coordinate system.
S5 carries out a cloud essence registration, as shown in Fig. 2, specifically comprising the following steps:
(1) it is obtained by S4 with the measurement pointcloud and model point cloud under robot base coordinate sys-tem；The condition of convergence, tool are set
Body includes threshold rotating value e, translation threshold value t and maximum number of iterations Nmax-itr；Kd-tree is established to model point cloud, for most
Neighbour's point search；
(2) each point p in measurement pointcloud is traversedi, according in kd-tree search model point cloud with it apart from nearest point qi
As match point, to obtain multipair matching double points, matching double points collection { C is formedi}；
(3) { C is calculatediIn all matching double points distance between two points average valueAnd standard deviation, work as match point
Centering distance between two pointsMeet(wherein s is the error scale factor), then retain the matching double points；It is no
Then, by it from matching double points collection { CiIn reject, obtain new matching double points collection { Ci', and reject the correspondence in measurement pointcloud
Point；
(4) preferably according to distance between two points in matching double pointsSize, by { Ci' in matching double points by it is small to
Big sequence, according to required matching double points quantity, selectionLesser part matching double points constitute matching double points collection { Ci″}；
(5) it uses to cloud noise and the better linear solution point of external acnode interference free performance to face iteration with regard to proximal method
(ICP), objective function is constructed:
Wherein, piWith qiIt is matching double points collection { Ci" in a pair of of point, and piFor the point in measurement pointcloud, qiFor model points
Point in cloud, niIt is qiNormal vector, NCIt is matching double points collection { Ci" in matching double points quantity；
Specifically, its rotation angle is θ when iteration updates measurement pointcloud, the rotation angle along three axis distinguishes α, β, γ, can recognize
It is α, β, γ all close to 0, sin θ ≈ θ, cos θ ≈ 1, and has T=(tx, ty, tz)T,
Formula (5) can then be converted are as follows:
Wherein,p″1
Indicate the coordinate at measurement pointcloud midpoint in first matching double points, n "1Indicate model point cloud midpoint in first matching double points
Normal vector,Indicate the coordinate at measurement pointcloud midpoint in the last one matching double points,Indicate the last one matching double points
The normal vector at middle model point cloud midpoint, q "1Indicate the coordinate at model point cloud midpoint in first matching double points,Indicate last
The coordinate at model point cloud midpoint in one matching double points；
Then α, β, γ, t are solved by the method for singular value decompositionx、ty、tz, and then R, T are acquired, and according to R, T couple
Measurement pointcloud carries out rigid transformation and obtains new measurement pointcloud；
(6) when iteration the cosine value cos θ > e and measurement pointcloud of the rotation angle of measurement pointcloud translation distance D < t
When or the number of iterations reach maximum number of iterations Nmax-itrWhen (maximum number of iterations is preferably 30 times), stop iteration, obtain most
Whole spin matrix R and translation matrix T completes point cloud essence registration；Otherwise the step (2) then repeated in S5 arrives (5), until meeting
The condition of convergence；
Wherein, according to spin matrix R define in angle-axis representation, obtainr11
Indicate first element of leading diagonal in spin matrix R, r22Indicate second element of leading diagonal in spin matrix R, r33
Indicate the third element of leading diagonal in spin matrix R.
As it will be easily appreciated by one skilled in the art that the foregoing is merely illustrative of the preferred embodiments of the present invention, not to
The limitation present invention, any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should all include
Within protection scope of the present invention.
Claims (6)
1. a kind of point cloud registration method based on hand and eye calibrating, which is characterized in that specifically comprise the following steps:
S1 measuring device is mounted on robot end, and is driven by robot end from initial position and be moved to the first measurement point,
Obtain transformation matrix A of robot end's coordinate system from initial position to the first measurement point1And measuring device coordinate system is from first
Beginning position to the first measurement point transformation matrix B1；
S2 robot end drives measuring device to be moved to the second measurement point from the first measurement point, obtains robot end's coordinate system
From the transformation matrix A of the first measurement point to the second measurement point2And measuring device coordinate system is measured from the first measurement point to second
The transformation matrix B of point2；
S3 is according to hand and eye calibrating equation, by the transformation matrix A1、B1、A2、B2Same position robot end's coordinate system is acquired to arrive
The transformation matrix of measuring device coordinate system；
S4 is sat by the transformation matrix of robot end's coordinate system to measuring device coordinate system and known robot base
Mark system arrive robot end's coordinate system transformation matrix, obtain robot base coordinate sys-tem to measuring device coordinate system transformation square
Battle array；Then according to this transformation matrix, the object under test measurement pointcloud that measuring device obtains is transformed into from measuring device coordinate system
In robot base coordinate sys-tem, the point cloud rough registration between measurement pointcloud and the model point cloud of object under test is completed；
S5 is obtained with the matching double points in the measurement pointcloud and model point cloud under robot base coordinate sys-tem, according to matching double points
Between relationship complete point cloud essence registration.
2. as described in claim 1 based on the point cloud registration method of hand and eye calibrating, which is characterized in that hand and eye calibrating in the S3
Equation specifically:
Wherein, X is transformation matrix of robot end's coordinate system to measuring device coordinate system.
3. as claimed in claim 1 or 2 based on the point cloud registration method of hand and eye calibrating, which is characterized in that use Matrix Direct Product
Method solves hand and eye calibrating equation.
4. as described in claim 1 based on the point cloud registration method of hand and eye calibrating, which is characterized in that the S5 specifically include as
Lower step:
(1) for each point in model point cloud, its nearest neighbor point is searched in measurement pointcloud respectively as match point, to obtain
Multipair matching double points are obtained, matching double points collection is formed；
(2) average value of distance between all matching double points is calculatedWith standard deviation, when distance between two points in matching double pointsIt is full
FootThis matching double points is then retained in matching double points to concentrate, is otherwise rejected, to update match point
To collection；Wherein, s is the error scale factor；
(3) objective function is constructed:
Wherein, piWith qiIt is a pair of of point that matching double points are concentrated, and piFor the point in measurement pointcloud, qiFor the point in model point cloud,
niIt is qiNormal vector, NCIt is the quantity that matching double points concentrate matching double points；To acquire the rotation of measurement pointcloud to model point cloud
Matrix R and translation matrix T, and measurement pointcloud is updated according to this spin matrix R and translation matrix T；
(4) repeat step (1) and arrive (3), when meeting the preset condition of convergence, stop iteration, obtain final spin matrix and
Translation matrix completes point cloud essence registration.
5. as claimed in claim 4 based on the point cloud registration method of hand and eye calibrating, which is characterized in that the step (2) updates
After matching double points collection, retain distance between two points in matching double pointsLesser part matching double points, update matching double points again
Collection.
6. as claimed in claim 4 based on the point cloud registration method of hand and eye calibrating, which is characterized in that the condition of convergence are as follows:
When iteration, the cosine value of measurement pointcloud rotation angle is greater than preset threshold rotating value and the translation distance of measurement pointcloud is less than in advance
If translation threshold value when or the number of iterations when reaching preset maximum number of iterations, objective function convergence stops iteration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910544860.4A CN110335296A (en) | 2019-06-21 | 2019-06-21 | A kind of point cloud registration method based on hand and eye calibrating |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910544860.4A CN110335296A (en) | 2019-06-21 | 2019-06-21 | A kind of point cloud registration method based on hand and eye calibrating |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110335296A true CN110335296A (en) | 2019-10-15 |
Family
ID=68142805
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910544860.4A Pending CN110335296A (en) | 2019-06-21 | 2019-06-21 | A kind of point cloud registration method based on hand and eye calibrating |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110335296A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110930442A (en) * | 2019-11-26 | 2020-03-27 | 广东技术师范大学 | Method and device for determining positions of key points in robot hand-eye calibration based on calibration block |
| CN112082483A (en) * | 2020-09-09 | 2020-12-15 | 易思维（杭州）科技有限公司 | Positioning method and application of object with edge characteristics only and precision evaluation method |
| CN112634435A (en) * | 2020-12-17 | 2021-04-09 | 中国地质大学（武汉） | Ceramic product three-dimensional point cloud reconstruction method based on Eye in Hand model |
| CN113558766A (en) * | 2021-07-19 | 2021-10-29 | 北京纳通医学研究院有限公司 | Image registration method and device, surgical robot and surgical robot system |
| CN113865506A (en) * | 2021-09-09 | 2021-12-31 | 武汉惟景三维科技有限公司 | Automatic three-dimensional measurement method and system for non-mark point splicing |
| CN116117818A (en) * | 2023-02-28 | 2023-05-16 | 哈尔滨工业大学 | Robot sensor hand-eye calibration method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108986149A (en) * | 2018-07-16 | 2018-12-11 | 武汉惟景三维科技有限公司 | A kind of point cloud Precision Registration based on adaptive threshold |
| CN108994844A (en) * | 2018-09-26 | 2018-12-14 | 广东工业大学 | A kind of scaling method and device of sanding operation arm trick relationship |
| CN109202912A (en) * | 2018-11-15 | 2019-01-15 | 太原理工大学 | A method of objective contour point cloud is registrated based on monocular depth sensor and mechanical arm |
| CN109465822A (en) * | 2018-10-22 | 2019-03-15 | 江苏集萃微纳自动化***与装备技术研究所有限公司 | Based on 3D vision hand and eye calibrating method |
-
2019
- 2019-06-21 CN CN201910544860.4A patent/CN110335296A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108986149A (en) * | 2018-07-16 | 2018-12-11 | 武汉惟景三维科技有限公司 | A kind of point cloud Precision Registration based on adaptive threshold |
| CN108994844A (en) * | 2018-09-26 | 2018-12-14 | 广东工业大学 | A kind of scaling method and device of sanding operation arm trick relationship |
| CN109465822A (en) * | 2018-10-22 | 2019-03-15 | 江苏集萃微纳自动化***与装备技术研究所有限公司 | Based on 3D vision hand and eye calibrating method |
| CN109202912A (en) * | 2018-11-15 | 2019-01-15 | 太原理工大学 | A method of objective contour point cloud is registrated based on monocular depth sensor and mechanical arm |
Non-Patent Citations (1)
| Title |
|---|
| 何若涛: "面向机器人操作的三维物体检测与位姿估计", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110930442A (en) * | 2019-11-26 | 2020-03-27 | 广东技术师范大学 | Method and device for determining positions of key points in robot hand-eye calibration based on calibration block |
| CN110930442B (en) * | 2019-11-26 | 2020-07-31 | 广东技术师范大学 | Method and device for determining positions of key points in robot hand-eye calibration based on calibration block |
| WO2021103824A1 (en) * | 2019-11-26 | 2021-06-03 | 广东技术师范大学 | Key point position determining method and device in robot hand-eye calibration based on calibration block |
| CN112082483A (en) * | 2020-09-09 | 2020-12-15 | 易思维（杭州）科技有限公司 | Positioning method and application of object with edge characteristics only and precision evaluation method |
| CN112634435A (en) * | 2020-12-17 | 2021-04-09 | 中国地质大学（武汉） | Ceramic product three-dimensional point cloud reconstruction method based on Eye in Hand model |
| CN113558766A (en) * | 2021-07-19 | 2021-10-29 | 北京纳通医学研究院有限公司 | Image registration method and device, surgical robot and surgical robot system |
| CN113865506A (en) * | 2021-09-09 | 2021-12-31 | 武汉惟景三维科技有限公司 | Automatic three-dimensional measurement method and system for non-mark point splicing |
| CN113865506B (en) * | 2021-09-09 | 2023-11-24 | 武汉惟景三维科技有限公司 | Automatic three-dimensional measurement method and system without mark point splicing |
| CN116117818A (en) * | 2023-02-28 | 2023-05-16 | 哈尔滨工业大学 | Robot sensor hand-eye calibration method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110335296A (en) | A kind of point cloud registration method based on hand and eye calibrating | |
| CN104392426B (en) | A kind of no marks point three-dimensional point cloud method for automatically split-jointing of self adaptation | |
| CN104463894B (en) | Multi-view three-dimensional laser point cloud global optimization integral registration method | |
| CN104515478B (en) | A kind of automatic method for three-dimensional measurement of high-precision blade of aviation engine and system | |
| US8055466B2 (en) | Global calibration for stereo vision probe | |
| CN105551039B (en) | The scaling method and device of structural light three-dimensional scanning system | |
| EP3255609B1 (en) | A method of automatically identifying a sequence of marking points in 3d medical image | |
| CN105806315B (en) | Noncooperative target relative measurement system and measuring method based on active coding information | |
| CN110335297A (en) | A kind of point cloud registration method based on feature extraction | |
| US20080243416A1 (en) | Global calibration for stereo vision probe | |
| CN108133458A (en) | A kind of method for automatically split-jointing based on target object spatial point cloud feature | |
| CN107133989A (en) | A kind of 3 D scanning system parameter calibration method | |
| CN104484508B (en) | Optimizing method for noncontact three-dimensional matching detection of complex curved-surface part | |
| CN106780459A (en) | A kind of three dimensional point cloud autoegistration method | |
| CN104634248B (en) | Revolving shaft calibration method under binocular vision | |
| CN110260786A (en) | A kind of robot vision measuring system and its scaling method based on external trace | |
| CN109579695B (en) | Part measuring method based on heterogeneous stereoscopic vision | |
| CN105447856B (en) | Reference points matching method based on robot motion's parameter and feature vector | |
| CN109297436A (en) | Binocular line laser measurement in space benchmark calibration method | |
| CN105654483A (en) | Three-dimensional point cloud full-automatic registration method | |
| Ravishankar et al. | Automated inspection of aircraft parts using a modified ICP algorithm | |
| CN105574812A (en) | Multi-angle three-dimensional data registration method and device | |
| CN113269673B (en) | Three-dimensional point cloud splicing method based on standard ball frame | |
| CN109087355A (en) | The monocular camera pose measuring apparatus and method updated based on iteration | |
| CN109242774A (en) | A kind of flat-type part point cloud method based on hyperspace invariant features |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20191015 |
|
| WD01 | Invention patent application deemed withdrawn after publication |