CN112001955A - Point cloud registration method and system based on two-dimensional projection plane matching constraint - Google Patents
Point cloud registration method and system based on two-dimensional projection plane matching constraint Download PDFInfo
- Publication number
- CN112001955A CN112001955A CN202010859158.XA CN202010859158A CN112001955A CN 112001955 A CN112001955 A CN 112001955A CN 202010859158 A CN202010859158 A CN 202010859158A CN 112001955 A CN112001955 A CN 112001955A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- point
- registration
- dimensional
- matching
- 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
- 238000000034 method Methods 0.000 title claims abstract description 46
- 230000009466 transformation Effects 0.000 claims abstract description 29
- 239000011159 matrix material Substances 0.000 claims abstract description 19
- 238000013507 mapping Methods 0.000 claims abstract description 16
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 230000008030 elimination Effects 0.000 claims abstract description 6
- 238000003379 elimination reaction Methods 0.000 claims abstract description 6
- 230000036544 posture Effects 0.000 claims abstract description 5
- 238000004891 communication Methods 0.000 claims abstract description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 37
- 238000009616 inductively coupled plasma Methods 0.000 claims description 27
- 238000012545 processing Methods 0.000 claims description 8
- 230000003993 interaction Effects 0.000 claims description 4
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 8
- 241000283973 Oryctolagus cuniculus Species 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 5
- 238000000746 purification Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 238000004590 computer program Methods 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 238000002940 Newton-Raphson method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Images
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
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a point cloud registration method and a point cloud registration system based on two-dimensional projection plane matching constraint, wherein the point cloud registration method comprises the following steps: s100, loading two point clouds, and projecting the point clouds to a screen coordinate system according to the initial postures of the point clouds to generate two pieces of two-dimensional image data; s200, respectively extracting Affine-SIFT feature points and descriptors of two pieces of two-dimensional image data, and calculating a ratio of nearest neighbor to next nearest neighbor to obtain an initial matching point set; s300, carrying out error matching point elimination on the initial matching point set to obtain ransac _ match _ ASIFT; s400, reading a pixel point set coordinate in the ransac _ match _ ASIFT to obtain a mapping point set three-dimensional coordinate of the pixel point set coordinate in the point cloud; s500, solving a transformation matrix through the three-dimensional coordinates of the mapping point set to complete image pre-registration; and S600, iterating to obtain an accurate registration result. The point cloud registration comprises a display module, an image registration module, a preprocessing module and a point cloud registration module, and the four modules are sequentially in communication connection.
Description
Technical Field
The invention relates to the technical field of computer vision, in particular to a point cloud registration method and system based on two-dimensional projection plane matching constraint.
Background
Conventional data registration is mostly image registration. Image registration based on image gray scale and image registration based on image features are two types of registration methods which are common at present. The registration method based on the image gray level has the characteristics of improving estimation accuracy and robustness, the registration principle is to convert two images into gray level images, then establish similarity measurement between the gray level images, then seek the maximum or minimum transformation model parameters of the similarity measurement, and the seeking process mainly adopts a searching mode, and the algorithm has the defects of large calculated amount and low calculating speed. The image feature-based registration method greatly compresses image information, accelerates calculation speed, has certain illumination change resistance, mainly extracts and matches the significant features of two images, so the method is sensitive to the errors of feature extraction modes and feature matching. The method is generally divided into four steps, namely feature extraction, feature matching, calculation of transformation models and transformation parameters, coordinate transformation and interpolation. The image registration technology is mature, and has wide application in the fields of remote sensing image processing, target change monitoring, target identification, medical image analysis and the like.
Inspired by the registration of two-dimensional data, in the mid-eighties of the twentieth century, many scholars began to study the spatial registration of three-dimensional data, which benefits from the application of three-dimensional laser scanning systems.
Faugeras and Heber proposed the concept of a quadruple in 1986, applying the quadruple to spatial registration, and proposed a matching method between a point set and a point set. With this inspiring in 1987, Horn and Arun et al propose a method pstps (point to point registration) for point set registration based on a four-tuple method, which was later proven to be a key method for point set registration. Besl and Mckay, the field of computer vision research later, introduced an iterative closest point method, namely an ICP algorithm, which is a high-level method based on a free form surface.
The ICP algorithm well solves the registration problem among complex point sets, but the traditional ICP algorithm has the defects that if the initial corresponding point set is not well selected, the initial corresponding point set is easy to fall into a local extreme value, each iteration needs to calculate the corresponding points of the target point set in the reference point set, the calculation speed is low, and then a plurality of scholars propose some improvement schemes.
In 1991, Chen calculates the tangent plane distance from a point in a first view to a point in a second view by using a least square method, and further obtains coordinate transformation. Besl provides a rapid ICP algorithm in 1992, the algorithm well solves the problem of solving a 6-degree-of-freedom data point transformation model, but the processing speed is still slow due to the fact that the shortest distance between a point and the model needs to be calculated; in 2001, Fan et al calculate the shortest distance from a registration point set to a reference point set by a method of solving a nonlinear equation set by a Newton-Raphson method, and then use a least square method to calculate a rotation matrix R and a translation vector t, but the algorithm only uses a model smaller than a data point set; wang proposes an algorithm for extracting a point cloud boundary by using a mapping principle from a two-dimensional image to a three-dimensional point cloud. In the aspect of point cloud registration, Wang Dong et al use a method of mapping two-dimensional SIFT feature points to three-dimensional corresponding points to complete point cloud registration, but the method is not applicable to point cloud registration with obvious affine transformation, Zhaka proposes a data fusion method based on k-domain information, Rusu uses the fast histogram feature of point cloud to solve the problem of point cloud pre-registration, Horn proposes the concept of Extended Gaussian Image (EGI) as early as 1984 and summarizes relevant properties thereof, Makadia uses a method of converting point cloud into Extended Gaussian Image (EGI) to complete point cloud full-automatic registration by using affine invariance thereof, is inspired by two-dimensional image registration, Cyr, Wang et al propose a 2D-3D mapping method, and uses 2D features as shape features and point features to obtain better pre-registration effect. In the aspect of surface reconstruction, Fang explains the Delaunay triangulation principle, and a Particle Swarm Optimization (PSO) is applied to the region growing method in Chengjinui, so that the triangulation efficiency is improved. Cheng and Huang et al propose some improved algorithms based on the ICP algorithm in terms of surface reconstruction and package them in PCL library.
The point cloud registration algorithm based on the local descriptor can automatically and accurately register point clouds with partial overlapping, however, when the point clouds to be registered have strong similar characteristics, a large number of mismatching points occur, which is not beneficial to accurately estimating an initial transformation matrix, and therefore, a point cloud registration method capable of reducing the mismatching points needs to be researched to solve the problems.
Disclosure of Invention
In order to solve the technical problems, the invention provides a point cloud registration method and a point cloud registration system based on two-dimensional projection plane matching constraint, the method and the system have the characteristic of affine transformation resistance, and the problem of establishing correct corresponding point sets of two point clouds with affine transformation when the two point clouds are shot at different view angles is well solved.
In order to achieve the purpose, the technical scheme of the invention is as follows:
a point cloud registration method based on two-dimensional projection plane matching constraint comprises the following steps:
s100, loading two point clouds, and projecting the point clouds to a screen coordinate system according to the initial postures of the point clouds to generate two pieces of two-dimensional image data;
s200, extracting Affinine-SIFT feature points and descriptors of two pieces of two-dimensional image data respectively by utilizing an Affinine-SIFT algorithm, and calculating a ratio of nearest neighbor to next nearest neighbor by utilizing an Euclidean distance to obtain an initial matching point set;
s300, carrying out error matching point elimination on the initial matching point set by utilizing a RANSAC algorithm to obtain RANSAC _ match _ ASIFT;
s400, reading a pixel point set coordinate in the ransac _ match _ ASIFT in the VTK, and obtaining a mapping point set three-dimensional coordinate of the pixel point set coordinate in the point cloud in a man-machine interaction mode;
s500, solving a transformation matrix by using a least square method through mapping the three-dimensional coordinates of the point set to complete image pre-registration;
and S600, iterating the pre-registered point cloud by utilizing an ICP (inductively coupled plasma) algorithm to obtain an accurate registration result.
Further, the step S100 includes the steps of:
s110, loading and displaying the two point clouds;
and S120, projecting the point cloud initial posture to a screen coordinate system to generate a two-dimensional plane image, and intercepting and storing the two-dimensional plane image as two-dimensional image data in a png format.
Further, the step S500 includes the steps of:
s510, manually selecting six groups of corresponding point sets from the three-dimensional coordinates of the mapping point sets to obtain a transformation matrix by using a least square method;
and S510, multiplying the transformation matrix by the point cloud to be registered to obtain a pre-registration result of the point cloud data.
Further, in step S600, an ICP algorithm is used to iterate the pre-registered point cloud for 4 times, so as to obtain an accurate registration result.
A point cloud registration system based on two-dimensional projection plane matching constraint comprises a display module for displaying point cloud data and two-dimensional image data, an image registration module for extracting feature points of the two-dimensional image data and completing initial matching of the feature points, a preprocessing module for purifying an initial matching point set and finding a corresponding mapping point set in a point cloud for the purified initial matching point set, and a point cloud registration module for solving a transformation matrix to complete pre-registration and performing ICP (inductively coupled plasma) iterative processing on the point cloud data after the pre-registration is completed; the display module, the image registration module, the preprocessing module and the point cloud registration module are sequentially in communication connection.
Compared with the prior art, the invention has the advantages and positive effects that:
according to the method, the Affine transformation resisting characteristic of Affinine-SIFT feature points of the two-dimensional image is utilized, meanwhile, the RANSAC algorithm is utilized to eliminate mismatching points of the initial matching point set, the two-dimensional feature points are mapped to obtain the corresponding point set of the three-dimensional point cloud, then the transformation matrix is obtained to obtain the pre-registration result, and then the ICP algorithm is utilized, so that the iteration times of the ICP algorithm are greatly reduced, the algorithm efficiency is improved, the condition that the ICP cannot converge due to the fact that the initial corresponding point set cannot be found is effectively prevented, the accurate registration result is obtained, and the calculation rate and the accuracy of point cloud registration are effectively improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a schematic flow diagram of the present invention;
FIG. 2 is a schematic diagram of the projection of a point cloud onto a two-dimensional plane;
FIG. 3 is an experimental sample diagram in an embodiment, which is a rabbit point cloud and a dragon point cloud at different viewing angles, respectively;
FIG. 4 is a two-dimensional image feature point matching result before and after error point elimination, wherein a is a matching result before rabbit point cloud RANSAC purification, and b is a matching result after rabbit point cloud RANSAC purification;
FIG. 5 is a two-dimensional image feature point matching result before and after error point elimination, wherein a is a matching result before dragon point cloud RANSAC purification, and b is a matching result after dragon point cloud RANSAC purification;
FIG. 6 is two sets of rabbit point clouds before and after registration, wherein a is the rabbit point cloud before registration, b is the rabbit point cloud after pre-registration, and c is the rabbit point cloud after ICP algorithm fine registration;
FIG. 7 is two sets of dragon point clouds before and after registration, wherein, a is the dragon point clouds before registration, b is the dragon point clouds after pre-registration, and c is the dragon point clouds after ICP algorithm fine registration;
fig. 8 is a frame structure diagram of the point cloud registration system.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived from the embodiments of the present invention by a person skilled in the art without any creative effort, should be included in the protection scope of the present invention.
In the embodiment, a corresponding computer program is written by using a C + + development language to automatically execute the invention, that is, the written computer program is used to complete the accurate registration of the point cloud.
In this embodiment, the rabbit point cloud and the dragon point cloud shown in fig. 3, 4, 5, 6, and 7 are taken as examples, and the written computer program is used to register point cloud data at two different viewing angles.
The flow of registration of the point cloud in this embodiment will be described in detail with reference to fig. 1.
And Step1, loading and storing the point cloud data.
Because the point cloud is three-dimensional data, the feature point extraction is directly carried out on the point cloud, the repetition rate is low and the point cloud is unstable, the point cloud data is loaded by using the system, a two-dimensional plane image (as shown in figure 2) is generated by projecting the initial attitude of the point cloud to a screen coordinate system, and the two-dimensional plane image is stored as two-dimensional image data in a PNG format.
Step2, processing of two-dimensional image data.
And extracting the characteristic points of the two-dimensional image data, and acquiring initial corresponding point set coordinates.
One specific embodiment of this step will be provided below.
There are many existing methods for extracting feature points, such as SIFT, ORB, SURF, etc., and when the front of the camera is blank, the direction of the optical axis of the camera may change, which causes distortion. Algorithms such as SIFT (scale invariant feature transform) have complete scale invariance but not complete affine invariance, and have certain limitation on the extraction of image features with large angle space change of shooting angles. In the embodiment, the Affinine-SIFT is used for extracting the feature points. The Affine-SIFT algorithm is high in automation, has Affine invariance and does not need manual intervention, and the method comprises the following specific steps:
firstly, selecting sampling parameters and simulating images with different longitudes and latitudes; then calculating the characteristics of the simulated image; and finally, combining the characteristics of all the simulated images to perform characteristic matching to obtain an initial characteristic point set.
Step3, processing of the initial feature point set.
The initial feature point set may include a plurality of erroneous point sets, if the number of erroneous point sets is too large, the final point cloud registration is affected, and the initial feature point set needs to be subjected to erroneous elimination, that is, point set purification. The RANSAC algorithm can eliminate the error matching of the point set to a large extent.
Step4, reading and mapping the feature point set.
The points calculated by the Affinine-SITF algorithm are two-dimensional image points, the point cloud data are three-dimensional image points, and three-dimensional coordinate points which are mapped one by one are found on the point cloud data for the two-dimensional image points to realize the point cloud registration. On the premise of consistent viewpoints and no scaling, the corresponding points of the two-dimensional pixel points in the three-dimensional point cloud are obtained by using a VTK-based human-computer interaction mode, so that the process of mapping from two dimensions to three dimensions is completed.
Step5, solving the transformation matrix.
Any one of the two point cloud data with different visual angles can be completely overlapped with the other point cloud data through operations of translation, scaling, rotation and the like, namely, each three-dimensional point of the two point clouds with different visual angles has a certain mathematical relationship with a corresponding point in the other point cloud, and the mathematical relationship is a transformation matrix between images. The method comprises the steps of obtaining corresponding points of two-dimensional pixel points in three-dimensional point cloud through a man-machine interaction mode based on VTK, manually selecting six groups of corresponding point sets from the points, obtaining a transformation matrix through a least square method, and multiplying the transformation matrix by point cloud to be registered to obtain a coarse registration result of point cloud data.
The selection of the corresponding point set in this embodiment is specifically as follows:
after the corresponding point set in the three-dimensional point cloud is obtained, six groups of point cloud corresponding points are manually selected by eyes, and the selected corresponding points cannot be excessively concentrated in the point cloud and are distributed in a balanced manner as far as possible.
Step6, coarse registration point cloud optimization processing.
Due to the fact that a corresponding point set is selected by manual visual observation to calculate a transformation matrix, selection errors may exist, a small amount of registration errors still exist in the point cloud subjected to coarse registration, deep registration needs to be conducted on the point cloud subjected to coarse registration, registration errors are further reduced, iteration is conducted by means of an ICP algorithm, and convergence can be completed only by iterating for 4 times. Under the condition that the ICP algorithm can conduct direct registration, the result can be converged only by iterating 30-50 times generally, and after coarse registration, the iteration times are reduced to 4, so that the efficiency of the ICP algorithm is greatly improved.
The invention provides a point cloud registration system for implementing the steps, as shown in fig. 8, the system comprises a display module for displaying point cloud data and two-dimensional image data, an image registration module for extracting feature points of the two-dimensional image data and completing initial matching of the feature points, a preprocessing module for purifying an initial matching point set and finding a corresponding mapping point set in the point cloud for the purified initial matching point set, and a point cloud registration module for solving a transformation matrix to complete pre-registration and performing ICP iterative processing on the point cloud data after the pre-registration is completed; the display module, the image registration module, the preprocessing module and the point cloud registration module are sequentially in communication connection.
The point cloud registration system can be implemented in electronic hardware, computer software, or a combination of both, and the components and steps of the examples have been generally described in the foregoing description in terms of functionality in order to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The steps of a method or algorithm described in this disclosure may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in random access memory, read only memory, electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
According to the method, the Affine transformation resisting characteristic of Affinine-SIFT feature points of the two-dimensional image is utilized, meanwhile, the RANSAC algorithm is utilized to eliminate mismatching points of the initial matching point set, the two-dimensional feature points are mapped to obtain the corresponding point set of the three-dimensional point cloud, then the transformation matrix is obtained to obtain the pre-registration result, and then the ICP algorithm is utilized, so that the iteration times of the ICP algorithm are greatly reduced, the algorithm efficiency is improved, the condition that the ICP cannot converge due to the fact that the initial corresponding point set cannot be found is effectively prevented, the accurate registration result is obtained, and the calculation rate and the accuracy of point cloud registration are effectively improved.
Claims (5)
1. A point cloud registration method based on two-dimensional projection plane matching constraint comprises the following steps:
s100, loading two point clouds, and projecting the point clouds to a screen coordinate system according to the initial postures of the point clouds to generate two pieces of two-dimensional image data;
s200, extracting Affinine-SIFT feature points and descriptors of two pieces of two-dimensional image data respectively by utilizing an Affinine-SIFT algorithm, and calculating a ratio of nearest neighbor to next nearest neighbor by utilizing an Euclidean distance to obtain an initial matching point set;
s300, carrying out error matching point elimination on the initial matching point set by utilizing a RANSAC algorithm to obtain RANSAC _ match _ ASIFT;
s400, reading a pixel point set coordinate in the ransac _ match _ ASIFT in the VTK, and obtaining a mapping point set three-dimensional coordinate of the pixel point set coordinate in the point cloud in a man-machine interaction mode;
s500, solving a transformation matrix by using a least square method through mapping the three-dimensional coordinates of the point set to complete image pre-registration;
and S600, iterating the pre-registered point cloud by utilizing an ICP (inductively coupled plasma) algorithm to obtain an accurate registration result.
2. The point cloud registration method based on two-dimensional projection plane matching constraints of claim 1, wherein: the step S100 includes the steps of:
s110, loading and displaying the two point clouds;
and S120, projecting the point cloud initial posture to a screen coordinate system to generate a two-dimensional plane image, and intercepting and storing the two-dimensional plane image as two-dimensional image data in a png format.
3. The point cloud registration method based on two-dimensional projection plane matching constraints of claim 2, wherein: the step S500 includes the steps of:
s510, manually selecting six groups of corresponding point sets from the three-dimensional coordinates of the mapping point sets to obtain a transformation matrix by using a least square method;
and S510, multiplying the transformation matrix by the point cloud to be registered to obtain a pre-registration result of the point cloud data.
4. The point cloud registration method based on two-dimensional projection plane matching constraints of claim 3, wherein: in the step S600, the ICP algorithm is used to iterate the pre-registered point cloud for 4 times, so as to obtain an accurate registration result.
5. A point cloud registration system based on two-dimensional projection plane matching constraint, the point cloud registration system being configured to implement the point cloud registration method based on two-dimensional projection plane matching constraint of claim 4, wherein: the point cloud registration system comprises a display module for displaying point cloud data and two-dimensional image data, an image registration module for extracting characteristic points of the two-dimensional image data and completing initial matching of the characteristic points, a preprocessing module for purifying an initial matching point set and finding a corresponding mapping point set in a point cloud for the purified initial matching point set, and a point cloud registration module for solving a transformation matrix to complete pre-registration and performing ICP (inductively coupled plasma) iterative processing on the point cloud data after the pre-registration is completed; the display module, the image registration module, the preprocessing module and the point cloud registration module are sequentially in communication connection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010859158.XA CN112001955A (en) | 2020-08-24 | 2020-08-24 | Point cloud registration method and system based on two-dimensional projection plane matching constraint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010859158.XA CN112001955A (en) | 2020-08-24 | 2020-08-24 | Point cloud registration method and system based on two-dimensional projection plane matching constraint |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112001955A true CN112001955A (en) | 2020-11-27 |
Family
ID=73470713
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010859158.XA Pending CN112001955A (en) | 2020-08-24 | 2020-08-24 | Point cloud registration method and system based on two-dimensional projection plane matching constraint |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112001955A (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112884902A (en) * | 2021-03-17 | 2021-06-01 | 中山大学 | Point cloud registration-oriented target ball position optimization method |
| CN113284170A (en) * | 2021-05-26 | 2021-08-20 | 北京智机科技有限公司 | Point cloud rapid registration method |
| CN113345023A (en) * | 2021-07-05 | 2021-09-03 | 北京京东乾石科技有限公司 | Positioning method and device of box body, medium and electronic equipment |
| CN113344986A (en) * | 2021-08-03 | 2021-09-03 | 深圳市信润富联数字科技有限公司 | Point cloud registration result evaluation method, device, equipment and storage medium |
| CN113450395A (en) * | 2021-06-16 | 2021-09-28 | 北京理工大学 | Surgical navigation space registration method and device based on optical tracking probe point drawing |
| CN113516606A (en) * | 2021-03-05 | 2021-10-19 | 河海大学 | Slope radar image and optical photo matching fusion method assisted by oblique photography data |
| 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 |
| CN113689476A (en) * | 2021-08-26 | 2021-11-23 | 中国科学院植物研究所 | A point cloud registration method, device, storage medium and computer equipment |
| CN113793370A (en) * | 2021-01-13 | 2021-12-14 | 北京京东叁佰陆拾度电子商务有限公司 | Three-dimensional point cloud registration method and device, electronic equipment and readable medium |
| CN113808201A (en) * | 2021-08-06 | 2021-12-17 | 亿嘉和科技股份有限公司 | Target object detection method and guided grabbing method |
| CN113870326A (en) * | 2021-09-01 | 2021-12-31 | 浙江大学 | A Structural Damage Mapping, Quantification and Visualization Method Based on Image and 3D Point Cloud Registration |
| CN114170565A (en) * | 2021-11-09 | 2022-03-11 | 广州市鑫广飞信息科技有限公司 | A method, device and terminal equipment for image comparison based on UAV aerial photography |
| CN114241018A (en) * | 2021-12-09 | 2022-03-25 | 电子科技大学 | A kind of tooth point cloud registration method, system and readable storage medium |
| CN114494368A (en) * | 2021-12-20 | 2022-05-13 | 东南大学 | Low-overlapping-rate point cloud registration method combining dimensionality reduction projection and feature matching |
| CN114708413A (en) * | 2022-03-31 | 2022-07-05 | 北京师范大学 | Corresponding point matching method of heart 3D model based on contour alignment |
| WO2022237225A1 (en) * | 2021-05-08 | 2022-11-17 | 广东三姆森科技股份有限公司 | Online real-time registration method for incomplete three-dimensional scanning point cloud having plane reference |
| CN116008177A (en) * | 2022-12-07 | 2023-04-25 | 广州市柯洱斯电子有限责任公司 | SMT component high defect identification method, system and readable medium thereof |
| CN116823895A (en) * | 2023-06-25 | 2023-09-29 | 东南大学 | Digital image calculation method and system for multi-view matching of RGB-D cameras based on variable templates |
| CN117974747A (en) * | 2024-04-01 | 2024-05-03 | 北京理工大学长三角研究院(嘉兴) | 360-Degree point cloud 2D depth disk triangulation composition method, device and system |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101320341A (en) * | 2007-06-07 | 2008-12-10 | 国际商业机器公司 | Systems and methods for recovery from hardware access errors |
| US20080310757A1 (en) * | 2007-06-15 | 2008-12-18 | George Wolberg | System and related methods for automatically aligning 2D images of a scene to a 3D model of the scene |
| CN102902596A (en) * | 2012-09-29 | 2013-01-30 | 北京百度网讯科技有限公司 | Point of interest data verification method and point of interest data verification device |
| CN103106688A (en) * | 2013-02-20 | 2013-05-15 | 北京工业大学 | Indoor three-dimensional scene rebuilding method based on double-layer rectification method |
| US20140003705A1 (en) * | 2012-06-29 | 2014-01-02 | Yuichi Taguchi | Method for Registering Points and Planes of 3D Data in Multiple Coordinate Systems |
| CN103744086A (en) * | 2013-12-23 | 2014-04-23 | 北京建筑大学 | High-precision registration method for ground laser radar and close-range photography measurement data |
| CN103843284A (en) * | 2013-10-23 | 2014-06-04 | 华为技术有限公司 | Disaster tolerance method, system and device of cloud application |
| CN105654422A (en) * | 2015-12-23 | 2016-06-08 | 北京观著信息技术有限公司 | Point cloud registration method and system |
| CN106780459A (en) * | 2016-12-12 | 2017-05-31 | 华中科技大学 | A kind of three dimensional point cloud autoegistration method |
| CN111062988A (en) * | 2019-11-29 | 2020-04-24 | 佛山科学技术学院 | A Grape Pose Estimation Method Based on Local Point Cloud |
-
2020
- 2020-08-24 CN CN202010859158.XA patent/CN112001955A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101320341A (en) * | 2007-06-07 | 2008-12-10 | 国际商业机器公司 | Systems and methods for recovery from hardware access errors |
| US20080310757A1 (en) * | 2007-06-15 | 2008-12-18 | George Wolberg | System and related methods for automatically aligning 2D images of a scene to a 3D model of the scene |
| US20140003705A1 (en) * | 2012-06-29 | 2014-01-02 | Yuichi Taguchi | Method for Registering Points and Planes of 3D Data in Multiple Coordinate Systems |
| CN102902596A (en) * | 2012-09-29 | 2013-01-30 | 北京百度网讯科技有限公司 | Point of interest data verification method and point of interest data verification device |
| CN103106688A (en) * | 2013-02-20 | 2013-05-15 | 北京工业大学 | Indoor three-dimensional scene rebuilding method based on double-layer rectification method |
| CN103843284A (en) * | 2013-10-23 | 2014-06-04 | 华为技术有限公司 | Disaster tolerance method, system and device of cloud application |
| CN103744086A (en) * | 2013-12-23 | 2014-04-23 | 北京建筑大学 | High-precision registration method for ground laser radar and close-range photography measurement data |
| CN105654422A (en) * | 2015-12-23 | 2016-06-08 | 北京观著信息技术有限公司 | Point cloud registration method and system |
| CN106780459A (en) * | 2016-12-12 | 2017-05-31 | 华中科技大学 | A kind of three dimensional point cloud autoegistration method |
| CN111062988A (en) * | 2019-11-29 | 2020-04-24 | 佛山科学技术学院 | A Grape Pose Estimation Method Based on Local Point Cloud |
Non-Patent Citations (5)
| Title |
|---|
| XU ZHANG: "Point cloud registration with 2D and 3D fusion information on mobile robot integrated vision system", 《2013 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND BIOMIMETICS (ROBIO)》 * |
| 刘博文: "基于二维特征点的三维人脸点云配准技术研究", 《中国优秀硕士论文全文数据库》, pages 20 - 38 * |
| 常江: "基于特征匹配的三维点云配准算法研究", 《中国优秀硕士论文全文数据库》 * |
| 沈泽皖: "基于ASIFT图像特征的大视角点云配准方法", 《实验力学》, pages 2 - 3 * |
| 马山庄;: "一种基于特征提取和匹配的点云自动配准算法", 科技创新与应用, no. 25 * |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113793370B (en) * | 2021-01-13 | 2024-04-19 | 北京京东叁佰陆拾度电子商务有限公司 | Three-dimensional point cloud registration method and device, electronic equipment and readable medium |
| CN113793370A (en) * | 2021-01-13 | 2021-12-14 | 北京京东叁佰陆拾度电子商务有限公司 | Three-dimensional point cloud registration method and device, electronic equipment and readable medium |
| CN113516606A (en) * | 2021-03-05 | 2021-10-19 | 河海大学 | Slope radar image and optical photo matching fusion method assisted by oblique photography data |
| CN113516606B (en) * | 2021-03-05 | 2023-12-12 | 河海大学 | Slope radar image and optical photo matching fusion method assisted by oblique photographing data |
| CN112884902A (en) * | 2021-03-17 | 2021-06-01 | 中山大学 | Point cloud registration-oriented target ball position optimization method |
| WO2022237225A1 (en) * | 2021-05-08 | 2022-11-17 | 广东三姆森科技股份有限公司 | Online real-time registration method for incomplete three-dimensional scanning point cloud having plane reference |
| CN113284170A (en) * | 2021-05-26 | 2021-08-20 | 北京智机科技有限公司 | Point cloud rapid registration method |
| CN113450395A (en) * | 2021-06-16 | 2021-09-28 | 北京理工大学 | Surgical navigation space registration method and device based on optical tracking probe point drawing |
| CN113450395B (en) * | 2021-06-16 | 2022-12-02 | 北京理工大学 | Surgical navigation space registration method and device based on optical tracking probe point drawing |
| CN113345023A (en) * | 2021-07-05 | 2021-09-03 | 北京京东乾石科技有限公司 | Positioning method and device of box body, medium and electronic equipment |
| CN113345023B (en) * | 2021-07-05 | 2024-03-01 | 北京京东乾石科技有限公司 | Box positioning method and device, medium and electronic equipment |
| 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 |
| CN113344986A (en) * | 2021-08-03 | 2021-09-03 | 深圳市信润富联数字科技有限公司 | Point cloud registration result evaluation method, device, equipment and storage medium |
| CN113808201A (en) * | 2021-08-06 | 2021-12-17 | 亿嘉和科技股份有限公司 | Target object detection method and guided grabbing method |
| CN113689476A (en) * | 2021-08-26 | 2021-11-23 | 中国科学院植物研究所 | A point cloud registration method, device, storage medium and computer equipment |
| CN113870326B (en) * | 2021-09-01 | 2023-11-28 | 浙江大学 | Structural damage mapping, quantifying and visualizing method based on image and three-dimensional point cloud registration |
| CN113870326A (en) * | 2021-09-01 | 2021-12-31 | 浙江大学 | A Structural Damage Mapping, Quantification and Visualization Method Based on Image and 3D Point Cloud Registration |
| CN114170565A (en) * | 2021-11-09 | 2022-03-11 | 广州市鑫广飞信息科技有限公司 | A method, device and terminal equipment for image comparison based on UAV aerial photography |
| CN114241018A (en) * | 2021-12-09 | 2022-03-25 | 电子科技大学 | A kind of tooth point cloud registration method, system and readable storage medium |
| CN114241018B (en) * | 2021-12-09 | 2023-05-09 | 电子科技大学 | A tooth point cloud registration method, system and readable storage medium |
| CN114494368A (en) * | 2021-12-20 | 2022-05-13 | 东南大学 | Low-overlapping-rate point cloud registration method combining dimensionality reduction projection and feature matching |
| CN114494368B (en) * | 2021-12-20 | 2024-11-08 | 东南大学 | A low overlap point cloud registration method combining dimensionality reduction projection and feature matching |
| CN114708413A (en) * | 2022-03-31 | 2022-07-05 | 北京师范大学 | Corresponding point matching method of heart 3D model based on contour alignment |
| CN114708413B (en) * | 2022-03-31 | 2024-11-29 | 北京师范大学 | Contour alignment-based corresponding point matching method for heart 3D model |
| CN116008177A (en) * | 2022-12-07 | 2023-04-25 | 广州市柯洱斯电子有限责任公司 | SMT component high defect identification method, system and readable medium thereof |
| CN116823895A (en) * | 2023-06-25 | 2023-09-29 | 东南大学 | Digital image calculation method and system for multi-view matching of RGB-D cameras based on variable templates |
| CN117974747A (en) * | 2024-04-01 | 2024-05-03 | 北京理工大学长三角研究院(嘉兴) | 360-Degree point cloud 2D depth disk triangulation composition method, device and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112001955A (en) | Point cloud registration method and system based on two-dimensional projection plane matching constraint | |
| CN110322500B (en) | Optimization method and device for instant positioning and map construction, medium and electronic equipment | |
| CN108052942B (en) | A visual image recognition method for aircraft flight attitude | |
| CN110111248B (en) | Image splicing method based on feature points, virtual reality system and camera | |
| CN107679537B (en) | A Pose Estimation Algorithm for Objects in Untextured Space Based on Contour Point ORB Feature Matching | |
| CN107833181B (en) | Three-dimensional panoramic image generation method based on zoom stereo vision | |
| CN108122256B (en) | A method of it approaches under state and rotates object pose measurement | |
| CN110070615A (en) | A kind of panoramic vision SLAM method based on polyphaser collaboration | |
| CN106548462B (en) | A Nonlinear SAR Image Geometric Correction Method Based on Thin Plate Spline Interpolation | |
| CN107358629B (en) | An indoor mapping and localization method based on target recognition | |
| CN107945217B (en) | A method and system for fast screening of image feature point pairs suitable for automatic assembly | |
| CN112053447A (en) | Augmented reality three-dimensional registration method and device | |
| CN107423772A (en) | A kind of new binocular image feature matching method based on RANSAC | |
| CN108416385A (en) | It is a kind of to be positioned based on the synchronization for improving Image Matching Strategy and build drawing method | |
| CN107527366B (en) | Camera tracking method for depth camera | |
| CN112419497A (en) | Monocular vision-based SLAM method combining feature method and direct method | |
| CN102446356A (en) | Parallel self-adaptive matching method for obtaining remote sensing images with uniformly distributed matching points | |
| CN114331879A (en) | Visible light and infrared image registration method for equalized second-order gradient histogram descriptor | |
| CN114266967A (en) | Cross-source remote sensing data target identification method based on symbolic distance characteristics | |
| CN117232525A (en) | Visual SLAM automatic navigation algorithm for palletizing robot | |
| CN116894876A (en) | 6-DOF positioning method based on real-time image | |
| CN110176041B (en) | A Novel Train Aided Assembly Method Based on Binocular Vision Algorithm | |
| CN116844124A (en) | Three-dimensional object detection frame labeling method, three-dimensional object detection frame labeling device, electronic equipment and storage medium | |
| CN110706332B (en) | Scene reconstruction method based on noise point cloud | |
| CN115423852A (en) | Point cloud data registration method and system, electronic device and storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201127 |