CN111915661B - Point cloud registration method, system and computer readable storage medium based on RANSAC algorithm - Google Patents
Point cloud registration method, system and computer readable storage medium based on RANSAC algorithm Download PDFInfo
- Publication number
- CN111915661B CN111915661B CN202010728477.7A CN202010728477A CN111915661B CN 111915661 B CN111915661 B CN 111915661B CN 202010728477 A CN202010728477 A CN 202010728477A CN 111915661 B CN111915661 B CN 111915661B
- Authority
- CN
- China
- Prior art keywords
- point cloud
- registration
- point
- matrix
- key surface
- 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
- 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
-
- 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
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Processing Or Creating Images (AREA)
Abstract
According to the method, the read point cloud to be registered is fitted through the RANSAC algorithm to obtain the key surface; extracting features according to the key surface to obtain a feature value; registering according to the characteristic values to obtain a rotation matrix and a translation matrix; and transforming the origin cloud coordinates according to the rotation matrix and the translation matrix to finish registration. The invention provides a point cloud registration method based on a RANSAC algorithm, which can effectively improve the registration speed on the premise of meeting the registration precision.
Description
Technical Field
The invention relates to the technical field of data analysis, in particular to a point cloud registration method, a system and a computer readable storage medium based on a RANSAC algorithm.
Background
The point cloud is a representation form of a three-dimensional object or a three-dimensional scene, and is composed of a group of irregularly distributed discrete points in space for expressing the spatial structure and surface properties of the three-dimensional object or the three-dimensional scene. And point cloud registration, namely, solving a rotation translation matrix between two point clouds, transforming a source point cloud (for example, a real-time map) to the same coordinate system as a target point cloud (for example, a preset map) so that the distance between the point cloud data in the two point clouds reaches a minimum value after the point cloud pose transformation.
The point cloud is a dense point set which accords with the measurement rule and can describe the surface characteristics of the target, is the third type of space data after vector and image, and can directly and effectively describe the three-dimensional real world. The method is mainly applied to the fields of mapping, automatic driving, planning and design, archaeology and cultural relics protection, medical treatment and the like. The point cloud registration is an important ring in the three-dimensional reconstruction process, and has a great influence on the overall accuracy after processing.
At present, the traditional point cloud registration algorithm needs to perform a large amount of operations, is low in efficiency, has no good initial pose of two-stage point cloud, is easy to sink into local optimum in point cloud registration, and has poor overall effect.
Disclosure of Invention
The invention provides a point cloud registration method, a system and a computer readable storage medium based on a RANSAC algorithm, which improve the efficiency and the accuracy of point cloud registration.
An embodiment of the present invention provides a point cloud registration method based on a RANSAC algorithm, including:
fitting the read point cloud to be registered through a RANSAC algorithm to obtain a key surface; wherein the key surface can represent an origin cloud;
extracting the characteristics of the key surface to obtain a characteristic value; wherein the characteristic value should be able to represent an origin cloud;
performing rough registration on the key surface according to the characteristic value to obtain a first rotation matrix and a translation matrix; adding the first rotation matrix and the translation matrix as parameters of fine registration;
carrying out fine registration on the key surface according to the parameters to obtain a second rotation matrix and a translation matrix;
and transforming the origin cloud coordinates according to the second rotation matrix and the translation matrix, and finishing registration.
Further, the method further comprises the following steps: and preprocessing the point cloud to be aligned, wherein the preprocessing comprises filtering and denoising.
Further, the method further comprises the following steps: and acquiring the point cloud to be registered.
One embodiment of the present invention provides a point cloud registration system based on a RANSAC algorithm, including:
the fitting module is used for fitting the read point cloud to be registered through a RANSAC algorithm to obtain a key surface; wherein the key surface can represent an origin cloud;
the characteristic value extraction module is used for extracting the characteristics of the key surface to obtain a characteristic value; wherein the characteristic value should be able to represent an origin cloud;
the rough registration module is used for carrying out rough registration on the key surface according to the characteristic value to obtain a first rotation matrix and a translation matrix; adding the first rotation matrix and the translation matrix as parameters of fine registration;
the fine registration module is used for carrying out fine registration on the key surface according to the parameters to obtain a second rotation matrix and a translation matrix;
and the coordinate transformation module is used for transforming the origin cloud coordinates according to the rotation matrix and the translation matrix to finish registration.
Further, the method further comprises the following steps: and the preprocessing module is used for preprocessing the point cloud to be aligned, wherein the preprocessing comprises filtering and denoising.
Further, the method further comprises the following steps: and the acquisition module is used for acquiring the point cloud to be registered.
An embodiment of the present invention provides a terminal for point cloud registration, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, where the processor implements the RANSAC algorithm-based point cloud registration method when executing the computer program.
One embodiment of the present invention provides a computer-readable storage medium comprising: the storage medium comprises a stored computer program, wherein the computer program is used for controlling equipment where the computer readable storage medium is located to execute the point cloud registration method based on the RANSAC algorithm when running.
Compared with the prior art, the embodiment of the invention has the beneficial effects that:
the method, the system and the computer readable storage medium for registering point clouds based on the RANSAC algorithm are provided, and the method fits the read point clouds to be registered through the RANSAC algorithm to obtain a key surface; extracting features according to the key surface to obtain a feature value; registering according to the characteristic values to obtain a rotation matrix and a translation matrix; and transforming the origin cloud coordinates according to the rotation matrix and the translation matrix to finish registration. The invention provides a point cloud registration method based on a RANSAC algorithm, which can effectively improve the registration speed on the premise of meeting the registration precision.
Drawings
In order to more clearly illustrate the technical solutions of the present invention, the drawings that are needed in the embodiments 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 that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a point cloud registration method based on a RANSAC algorithm according to an embodiment of the present invention;
fig. 2 is a flowchart of a point cloud registration method based on a RANSAC algorithm according to another embodiment of the present invention;
FIG. 3 is a flowchart of a point cloud registration method based on a RANSAC algorithm according to another embodiment of the present invention;
fig. 4 is a device diagram of a point cloud registration system based on a RANSAC algorithm according to an embodiment of the present invention;
fig. 5 is a device diagram of a point cloud registration system based on a RANSAC algorithm according to another embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that the step numbers used herein are for convenience of description only and are not limiting as to the order in which the steps are performed.
It is to be understood that the terminology used in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
The terms "comprises" and "comprising" indicate the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The term "and/or" refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
In a first aspect.
Referring to fig. 1, an embodiment of the present invention provides a point cloud registration method based on a RANSAC algorithm, including:
and S10, fitting the read point cloud to be registered through a RANSAC algorithm to obtain a key surface.
Wherein the key surface can represent an origin cloud.
Further, the read point clouds to be registered are fitted through a RANSAC algorithm, and three-dimensional point clouds are obtained, wherein the three-dimensional point clouds are key faces, and the number of the point clouds is reduced on the premise of meeting the condition that the original point cloud characteristics are not lost, so that registration is effectively quickened.
And S20, extracting the characteristics of the key surface to obtain a characteristic value.
Wherein the characteristic value should be able to represent an origin cloud; the feature extraction generally extracts normal vectors, curvatures and the like of the point cloud, but the number of geometric features around the point is large, the similarity is high, and global feature information of the point cloud cannot be obtained, so that the point feature histogram is used for making geometric description through the space difference between the point and the adjacent point, and the information provided by the point feature histogram has rotation invariance and is very robust to the point cloud.
Further, feature extraction is carried out on the key surface through a RANSAC algorithm, and feature values are obtained.
S30, performing rough registration on the key surface according to the characteristic value to obtain a first rotation matrix and a translation matrix; the first rotation matrix and the translation matrix are added as parameters of fine registration.
Further, the coarse registration algorithm is as follows:
1) N points are selected from the source point cloud B, and in order to ensure that the selected points have different point feature histograms, the distance between the selected points must be smaller than a given minimum threshold.
2) The target point cloud A searches points meeting similar conditions with the source point cloud B, and maintains a one-to-one correspondence.
3) Calculating a rotation matrix and a translation matrix of the corresponding points, and judging according to the Huber function
Wherein: m is a given threshold; li is the distance difference after the i-th group of corresponding points are transformed.
Repeating the steps until the result is optimal, namely, the error function takes the minimum value, and obtaining a translation matrix and a rotation matrix.
And S40, carrying out fine registration on the key surface according to the parameters to obtain a second rotation matrix and a translation matrix.
Further, the fine registration algorithm is as follows:
two-phase point clouds a and B, the point sets are a= { a1, a2, a3...an }, b= { B1, B2, b3...bm }, and after rotation-translation transformation, the points in the point clouds A, B are in one-to-one correspondence.
a i ＝R·b i +T
Wherein: r is a rotation matrix; t is the translation matrix.
Wherein the rotation matrix R and the translation matrix T are such that the objective function
Take the minimum value, at which point R, T is the optimal parameter.
9) Taking a point set ai from the target point cloud A, and finding a corresponding point bi from the source point cloud B to enable
a i -b i ＝min
10 A rotation matrix and a translation matrix are calculated so that the objective function takes a minimum value.
11 Rotating and translating the target point cloud A, and updating to obtain a new point cloud data set A'.
12 Calculating the distance between all the corresponding points in the updated point cloud A' and the source point cloud B, and performing normalization processing to obtain
Wherein: a' is to update and obtain a point set in the new point cloud data set; n is the number of new point cloud data concentrated points obtained by updating;
and (3) setting a threshold, repeating the steps if the average distance d is smaller than the given threshold, otherwise, regarding convergence.
Further, the key faces are fine registered by means of an icp algorithm (iterative closest point algorithm).
S50, converting the origin cloud coordinates according to the second rotation matrix and the translation matrix, and finishing registration.
Referring to fig. 2, an embodiment of the present invention provides a point cloud registration method based on a RANSAC algorithm, which further includes:
and S01, acquiring the point cloud to be registered.
S02, preprocessing the point cloud to be aligned; wherein the preprocessing includes filtering denoising.
Referring to fig. 3, in a specific implementation, an embodiment of the present invention provides a method for point cloud registration based on RANSAC algorithm, including:
and simultaneously acquiring point clouds A, B to be registered.
Preprocessing the point cloud A, B to be registered at the same time; wherein the preprocessing includes filtering denoising.
And simultaneously, fitting the read point cloud to be registered through a RANSAC algorithm to obtain a key surface A, B.
The key surface A, B can represent the original point cloud A, B to be aligned.
And simultaneously extracting the characteristics of the key surface A, B to obtain two groups of characteristic values.
Wherein the two sets of eigenvalues should represent an origin cloud A, B, respectively.
Performing coarse registration on the key surface according to the special two sets of characteristic values to obtain a first rotation matrix and a translation matrix; the first rotation matrix and the translation matrix are added as parameters of fine registration.
And carrying out fine registration on the key surface according to the parameters to obtain a second rotation matrix and a translation matrix.
And transforming the origin cloud coordinates according to the second rotation matrix and the translation matrix, and finishing registration.
The second aspect.
Referring to fig. 4, an embodiment of the present invention provides a point cloud registration system based on a RANSAC algorithm, including:
10. fitting a module; the method comprises the steps of fitting a read point cloud to be registered through a RANSAC algorithm to obtain a key surface; wherein the key surface can represent an origin cloud.
Further, the read point clouds to be registered are fitted through a RANSAC algorithm, and three-dimensional point clouds are obtained, wherein the three-dimensional point clouds are key faces, and the number of the point clouds is reduced on the premise of meeting the condition that the original point cloud characteristics are not lost, so that registration is effectively quickened.
20. The characteristic value extraction module; the method comprises the steps of extracting features of the key surface to obtain a feature value; wherein the characteristic value should be able to represent an origin cloud; the feature extraction generally extracts normal vectors, curvatures and the like of the point cloud, but the number of geometric features around the point is large, the similarity is high, and global feature information of the point cloud cannot be obtained, so that the point feature histogram is used for making geometric description through the space difference between the point and the adjacent point, and the information provided by the point feature histogram has rotation invariance and is very robust to the point cloud.
Further, feature extraction is carried out on the key surface through a RANSAC algorithm, and feature values are obtained.
30. A coarse registration module; the method comprises the steps of performing rough registration on a key surface according to the characteristic value to obtain a first rotation matrix and a translation matrix; the first rotation matrix and the translation matrix are added as parameters of fine registration.
Further, the coarse registration algorithm is as follows:
1) N points are selected from the source point cloud B, and in order to ensure that the selected points have different point feature histograms, the distance between the selected points must be smaller than a given minimum threshold.
2) The target point cloud A searches points meeting similar conditions with the source point cloud B, and maintains a one-to-one correspondence.
3) Calculating a rotation matrix and a translation matrix of the corresponding points, and judging according to the Huber function
Wherein: m is a given threshold; li is the distance difference after the i-th group of corresponding points are transformed.
Repeating the steps until the result is optimal, namely, the error function takes the minimum value, and obtaining a translation matrix and a rotation matrix.
40. A fine registration module; and the fine registration is used for carrying out fine registration on the key surface according to the parameters to obtain a second rotation matrix and a translation matrix.
Further, the fine registration algorithm is as follows:
two-phase point clouds a and B, the point sets are a= { a1, a2, a3...an }, b= { B1, B2, b3...bm }, and after rotation-translation transformation, the points in the point clouds A, B are in one-to-one correspondence.
a i ＝R·b i +T
Wherein: r is a rotation matrix; t is the translation matrix.
Wherein the rotation matrix R and the translation matrix T are such that the objective function
Take the minimum value, at which point R, T is the optimal parameter.
1) Taking a point set ai from the target point cloud A, and finding a corresponding point bi from the source point cloud B to enable
a i -b i ＝min
2) And calculating a rotation matrix and a translation matrix to enable the objective function to take the minimum value.
3) And carrying out rotation translation transformation on the target point cloud A, and updating to obtain a new point cloud data set A'.
4) Calculating the distance between all the corresponding points in the updated point cloud A' and the source point cloud B, and performing normalization processing to obtain
And (3) setting a threshold, repeating the steps if the average distance d is smaller than the given threshold, otherwise, regarding convergence.
50. A coordinate transformation module; the method is used for completing registration according to the origin cloud coordinates transformed by the rotation matrix and the translation matrix.
Referring to fig. 5, an embodiment of the present invention provides a point cloud registration system based on a RANSAC algorithm, which further includes:
01. an acquisition module; and the method is used for acquiring the point cloud to be registered. And the number of the point clouds to be registered is at least two.
02. A preprocessing module; the method comprises the steps of preprocessing the point cloud to be aligned, wherein the preprocessing comprises filtering and denoising.
In a third aspect.
An embodiment of the present invention provides a computer readable storage medium, which is characterized in that the computer readable storage medium includes a stored computer program, where when the computer program runs, a device where the computer readable storage medium is controlled to execute a point cloud registration method based on a RANSAC algorithm as set forth in any one of claims 1 to 3.
In another exemplary embodiment, a computer readable storage medium is also provided, comprising program instructions which, when executed by a processor, implement the steps of the above-described RANSAC algorithm based point cloud registration method. For example, the computer readable storage medium may be the above memory including program instructions executable by the processor to perform the above street lamp management method and achieve technical effects consistent with the above method.
While the foregoing is directed to the preferred embodiments of the present invention, it will be appreciated by those skilled in the art that changes and modifications may be made without departing from the principles of the invention, such changes and modifications are also intended to be within the scope of the invention.
Claims (8)
1. A point cloud registration method based on a RANSAC algorithm is characterized by comprising the following steps:
fitting the read point cloud to be registered through a RANSAC algorithm to obtain a key surface; wherein the key surface can represent an origin cloud;
extracting the characteristics of the key surface to obtain a characteristic value; wherein the characteristic value should be able to represent an origin cloud;
performing rough registration on the key surface according to the characteristic value to obtain a first rotation matrix and a translation matrix; adding the first rotation matrix and the translation matrix as parameters of fine registration;
carrying out fine registration on the key surface according to the parameters to obtain a second rotation matrix and a translation matrix;
further, the fine registration algorithm is as follows:
two-phase point clouds a and B, wherein the point sets are a= { a1, a2, a3...an }, b= { B1, B2, b3...bm }, and after rotation-translation transformation, the points in the point clouds A, B are in one-to-one correspondence;
a i ＝R·b i +T；
wherein: r is a rotation matrix; t is a translation matrix;
wherein the rotation matrix R and the translation matrix T are such that the objective function:
taking the minimum value, wherein R, T is the optimal parameter;
1) Taking a point set ai from the target point cloud A, and finding a corresponding point bi from the source point cloud B to enable
||a i -b i ||＝min；
2) Calculating a rotation matrix and a translation matrix to enable an objective function to take a minimum value;
3) Performing rotation translation transformation on the target point cloud A, and updating to obtain a new point cloud data set A';
4) Calculating the distances between all corresponding points in the updated point cloud A' and the source point cloud B, and carrying out normalization processing to obtain:
wherein: a' is to update and obtain a point set in the new point cloud data set; n is the number of new point cloud data concentrated points obtained by updating;
if the average distance d is smaller than the given threshold value, repeating the steps, otherwise, regarding convergence;
further, fine registration is performed on the key surface through an icp algorithm (iterative closest point algorithm);
and transforming the origin cloud coordinates according to the second rotation matrix and the translation matrix, and finishing registration.
2. The RANSAC algorithm-based point cloud registration method according to claim 1, wherein before the fitting of the key surface according to the read point cloud to be registered to the key surface, the method further comprises: and preprocessing the point cloud to be aligned, wherein the preprocessing comprises filtering and denoising.
3. The RANSAC algorithm-based point cloud registration method according to claim 1, wherein before the fitting of the key surface according to the read point cloud to be registered to the key surface, the method further comprises: and acquiring the point cloud to be registered.
4. A RANSAC algorithm-based point cloud registration system, comprising:
the fitting module is used for fitting the read point cloud to be registered through a RANSAC algorithm to obtain a key surface; wherein the key surface can represent an origin cloud;
the characteristic value extraction module is used for extracting the characteristics of the key surface to obtain a characteristic value; wherein the characteristic value should be able to represent an origin cloud;
the rough registration module is used for carrying out rough registration on the key surface according to the characteristic value to obtain a first rotation matrix and a translation matrix; adding the first rotation matrix and the translation matrix as parameters of fine registration;
the fine registration module is used for carrying out fine registration on the key surface according to the parameters to obtain a second rotation matrix and a translation matrix;
further, the fine registration algorithm is as follows:
two-phase point clouds a and B, wherein the point sets are a= { a1, a2, a3...an }, b= { B1, B2, b3...bm }, and after rotation-translation transformation, the points in the point clouds A, B are in one-to-one correspondence;
a i ＝R·b i +T；
wherein: r is a rotation matrix; t is a translation matrix;
wherein the rotation matrix R and the translation matrix T are such that the objective function:
taking the minimum value, wherein R, T is the optimal parameter;
5) Taking a point set ai from the target point cloud A, and finding a corresponding point bi from the source point cloud B to enable
||a i -b i ||＝min；
6) Calculating a rotation matrix and a translation matrix to enable an objective function to take a minimum value;
7) Performing rotation translation transformation on the target point cloud A, and updating to obtain a new point cloud data set A';
8) Calculating the distances between all corresponding points in the updated point cloud A' and the source point cloud B, and carrying out normalization processing to obtain:
wherein: a' is to update and obtain a point set in the new point cloud data set; n is the number of new point cloud data concentrated points obtained by updating;
if the average distance d is smaller than the given threshold value, repeating the steps, otherwise, regarding convergence;
further, fine registration is performed on the key surface through an icp algorithm (iterative closest point algorithm);
and the coordinate transformation module is used for transforming the origin cloud coordinates according to the rotation matrix and the translation matrix to finish registration.
5. The RANSAC algorithm-based point cloud registration system of claim 4, further comprising: and the preprocessing module is used for preprocessing the point cloud to be aligned, wherein the preprocessing comprises filtering and denoising.
6. The RANSAC algorithm-based point cloud registration system of claim 4, further comprising: and the acquisition module is used for acquiring the point cloud to be registered.
7. A terminal of point cloud registration, characterized by comprising a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, the processor implementing the RANSAC algorithm-based point cloud registration method according to any of claims 1 to 3 when executing the computer program.
8. A computer-readable storage medium, comprising: the storage medium comprising a stored computer program, wherein the computer program, when run, controls a device in which the computer readable storage medium is located to perform the RANSAC algorithm-based point cloud registration method according to any of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010728477.7A CN111915661B (en) | 2020-07-24 | 2020-07-24 | Point cloud registration method, system and computer readable storage medium based on RANSAC algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010728477.7A CN111915661B (en) | 2020-07-24 | 2020-07-24 | Point cloud registration method, system and computer readable storage medium based on RANSAC algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111915661A CN111915661A (en) | 2020-11-10 |
| CN111915661B true CN111915661B (en) | 2023-08-08 |
Family
ID=73280848
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010728477.7A Active CN111915661B (en) | 2020-07-24 | 2020-07-24 | Point cloud registration method, system and computer readable storage medium based on RANSAC algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111915661B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114485608B (en) * | 2021-12-13 | 2023-10-10 | 武汉中海庭数据技术有限公司 | Local point cloud rapid registration method for high-precision map making |
| CN114782672B (en) * | 2022-04-11 | 2024-06-21 | 清华大学 | Point cloud pose adjustment method and device |
| CN115063465B (en) * | 2022-06-15 | 2023-07-18 | 华南理工大学 | Unmanned vehicle driving road condition modeling method based on laser radar |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108564605A (en) * | 2018-04-09 | 2018-09-21 | 大连理工大学 | A kind of three-dimensional measurement spots cloud optimization method for registering |
| CN109272537A (en) * | 2018-08-16 | 2019-01-25 | 清华大学 | A kind of panorama point cloud registration method based on structure light |
| CN109919984A (en) * | 2019-04-15 | 2019-06-21 | 武汉惟景三维科技有限公司 | A kind of point cloud autoegistration method based on local feature description's |
| CN110930495A (en) * | 2019-11-22 | 2020-03-27 | 哈尔滨工业大学（深圳） | Multi-unmanned aerial vehicle cooperation-based ICP point cloud map fusion method, system, device and storage medium |
-
2020
- 2020-07-24 CN CN202010728477.7A patent/CN111915661B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108564605A (en) * | 2018-04-09 | 2018-09-21 | 大连理工大学 | A kind of three-dimensional measurement spots cloud optimization method for registering |
| CN109272537A (en) * | 2018-08-16 | 2019-01-25 | 清华大学 | A kind of panorama point cloud registration method based on structure light |
| CN109919984A (en) * | 2019-04-15 | 2019-06-21 | 武汉惟景三维科技有限公司 | A kind of point cloud autoegistration method based on local feature description's |
| CN110930495A (en) * | 2019-11-22 | 2020-03-27 | 哈尔滨工业大学（深圳） | Multi-unmanned aerial vehicle cooperation-based ICP point cloud map fusion method, system, device and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 吴建国 ; 程良伦 ; 白亚男 ; .基于3D-SIFT和区域生长的配准方法.中国科技信息.2018,(第05期),第1-3页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111915661A (en) | 2020-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111915661B (en) | Point cloud registration method, system and computer readable storage medium based on RANSAC algorithm | |
| CN110443836B (en) | Point cloud data automatic registration method and device based on plane features | |
| CN109559340B (en) | Parallel three-dimensional point cloud data automatic registration method | |
| CN109215129B (en) | Local feature description method based on three-dimensional point cloud | |
| CN108257155B (en) | Extended target stable tracking point extraction method based on local and global coupling | |
| CN107066969A (en) | A kind of face identification method | |
| CN112419497A (en) | Monocular vision-based SLAM method combining feature method and direct method | |
| CN115329111B (en) | Image feature library construction method and system based on point cloud and image matching | |
| CN112651944A (en) | 3C component high-precision six-dimensional pose estimation method and system based on CAD model | |
| CN111914832B (en) | SLAM method of RGB-D camera under dynamic scene | |
| CN116091727A (en) | Complex Qu Miandian cloud registration method based on multi-scale feature description, electronic equipment and storage medium | |
| CN111709893A (en) | ORB-SLAM2 improved algorithm based on information entropy and sharpening adjustment | |
| CN109766850B (en) | Fingerprint image matching method based on feature fusion | |
| CN116563582A (en) | Image template matching method and device based on domestic CPU and opencv | |
| CN114283199A (en) | Dynamic scene-oriented dotted line fusion semantic SLAM method | |
| Lin et al. | An Improved ICP with Heuristic Initial Pose for Point Cloud Alignment | |
| CN111860084B (en) | Image feature matching and positioning method and device and positioning system | |
| CN108917768B (en) | Unmanned aerial vehicle positioning navigation method and system | |
| Yu et al. | Shape analysis and recognition based on skeleton and morphological structure | |
| Makovetskii et al. | Registration algorithm for incongruent point clouds | |
| CN111445578A (en) | Map three-dimensional road feature identification method and system | |
| Yu et al. | Registration method for point clouds of complex rock mass based on dual structure information | |
| CN111127667A (en) | Point cloud initial registration method based on region curvature binary descriptor | |
| Sun et al. | Ncc feature matching optimized algorithm based on constraint fusion | |
| CN117671011B (en) | AGV positioning precision improving method and system based on improved ORB algorithm |
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 |