CN114972459A - Point cloud registration method based on low-dimensional point cloud local feature descriptor - Google Patents
Point cloud registration method based on low-dimensional point cloud local feature descriptor Download PDFInfo
- Publication number
- CN114972459A CN114972459A CN202210611042.3A CN202210611042A CN114972459A CN 114972459 A CN114972459 A CN 114972459A CN 202210611042 A CN202210611042 A CN 202210611042A CN 114972459 A CN114972459 A CN 114972459A
- Authority
- CN
- China
- Prior art keywords
- point
- point cloud
- descriptor
- neighborhood
- points
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
- G06T7/344—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
- G06T7/75—Determining position or orientation of objects or cameras using feature-based methods involving models
-
- 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)
- Processing Or Creating Images (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a point cloud registration method based on a low-dimensional point cloud local feature descriptor. Firstly, extracting key points of two pieces of point clouds through uniform sampling, and transferring the key points to a new 3D space to form a 3D descriptor by constructing a local reference coordinate system; then, combining the sum of included angles, curvatures, distances and normal vectors in the point cloud neighborhood to form a 'neighborhood point feature histogram' descriptor to encode the neighborhood information of the key points; the 3D descriptor and the neighborhood point feature histogram descriptor jointly form a low-dimensional point cloud local feature descriptor; the proposed descriptor firstly carries out radial search on the position of a 3D key point in a new 3D space, so that the search space of a corresponding point pair is reduced; and finally, registering by using a descriptor of 'neighborhood point feature histogram' through a RANSAC algorithm. The algorithm can obtain accurate registration effect in a short time, and is suitable for the field of precision measurement with high requirement on the registration effect.
Description
Technical Field
The invention belongs to the field of point cloud registration, and relates to a point cloud registration method for a low-dimensional point cloud local feature descriptor.
Background
With the development of computer technology, the three-dimensional modeling technology is increasingly widely applied in the fields of engineering modeling, digital cities, cultural relic protection, automatic driving and the like. How to obtain complete object surface information is one of the most fundamental topics. Due to the limited viewing angle of the point cloud collection device, the shielding of the object, the collection environment, and other limitations, usually, multi-angle and multi-batch collection is required to obtain the complete surface information of the object. The point cloud matching criterion is a process of integrating the point cloud data acquired from multiple batches and multiple angles into the same visual angle. Mathematically, each point cloud data has its own coordinate system. The problem to be solved by point cloud registration is to convert point cloud data of different coordinate systems into the same coordinate system through coordinate system transformation. The currently proposed point cloud registration algorithm usually needs to extract local features from a large amount of point cloud data, each local feature corresponds to a high-dimensional description vector, and a matching pair needs to be searched among a large number of point pairs, so that the problems of large calculation amount, low calculation efficiency and the like are caused. Therefore, aiming at the problems, the invention provides a point cloud registration method based on a low-dimensional point cloud local feature descriptor, which reduces the corresponding point search space by using the positions of key points in a new 3D space, and then finds accurate matching by using a descriptor of a neighborhood point feature histogram, thereby effectively improving the calculation speed and the accuracy.
Disclosure of Invention
The invention aims to solve the problems of high algorithm complexity, long program running time and low registration precision of the conventional mainstream three-dimensional point cloud registration algorithm, and provides a three-dimensional point cloud registration method based on a low-dimensional point cloud local feature descriptor. The invention transfers the points in the neighborhood of the key points through the 3D descriptor to a new 3D space, so that the key points in similar 3D surfaces are close to each other, and the search space is greatly reduced. And then, a 'neighborhood point feature histogram' descriptor is constructed by using the sum of the included angles of the normal vectors between the key points and the neighborhood points, the sum of curvature values, the distance values and three parameters. The 3D descriptor and the neighborhood point feature histogram descriptor jointly form a low-dimensional point cloud local feature descriptor. The provided descriptor has the characteristics of representativeness and high extraction speed, and the accuracy is improved while the time consumption of point cloud registration is greatly reduced.
The technical scheme of the invention is as follows: a point cloud registration algorithm based on a low-dimensional point cloud local feature descriptor is provided, and comprises the following basic steps:
step 1: acquiring point cloud data;
and 2, step: calculating the gravity center through voxel filtering, and traversing the nearest neighbor point of the gravity center as a key point;
and step 3: the keypoints are transferred to a new 3D space, whose new 3D coordinates are stored in the first three dimensions of the descriptor. A radial search within the radius of the keypoint is set in the new 3D space, retrieving a list of possible matches.
And 4, step 4: and constructing a three-dimensional feature descriptor by calculating normal vectors, curvatures and distances of key points and points in the radius neighborhood.
And 5: a radial nearest neighbor search is performed using Kd-Tree to find the exact nearest neighbors in the posterior three-dimensional descriptor space using euclidean metrics, and the exact match of the source keypoint under consideration is found using the sample consensus algorithm from the list of possible matches generated above.
The point clouds in the point cloud registration are divided into a source point cloud and a target point cloud. The source point cloud and the target point cloud are completely paired point clouds or point clouds collected by a laser triangular contourgraph, and are used for small-sized workpiece point clouds or outdoor large-sized point cloud data.
The step 2 specifically comprises the following steps:
the method equally divides the minimum bounding box surrounding the whole point cloud into a plurality of same small cubes, calculates the gravity center of each non-empty cube, and searches the adjacent points of the gravity center to replace the small cubes, thereby realizing the down-sampling and maintaining the original characteristic information.
The step 3 specifically comprises the following steps:
step 3.1: calculating a Key Point K i Surface radius r neighborhood space Surface ik Internal centroid coordinate mean pt ；
Step 3.2: surface of ik In all point coordinates minus centroid coordinate mean pt Effectively generate a new surface
Step 3.3: let the key point Ki subtract the centroid coordinate mean pt To obtain K i-meanpt ；
Step 3.4: we draw from a new surface
The modified covariance matrix COV is calculated as follows:
wherein: for readability, we use q m Represented on a 3D surface
Point (1) is represented by q
Point of (1), d m ＝‖q m -q‖ 2 。
Step 3.4: in order to create a unique local reference coordinate system and eliminate symbol ambiguity, the local x-axis and z-axis directions are oriented in most directions of the vector they represent, and finally, the local y-axis is obtained by the outer product of z and x, which is y ═ z × x.
Constructing a local reference coordinate system, described by the following formula:
[RF] 3×3 ＝[x T y T z T ] T
the keypoints are then transformed into the new 3D space, the keypoint coordinates are encoded as "3D" descriptors and a radius threshold is set to retrieve a list of possible matches, as follows:
the step 4 specifically comprises the following steps:
step 4.1: calculating a Key Point K i Normal phasor
And neighborhood space Surface ik Interior point normal vector
Included angle
Comprises the following steps:
wherein the content of the first and second substances,
the value range is [0, pi]。
The key point K i Normal phasor
And neighborhood space Surface ir Interior point normal vector
The sum of the included angles is:
and 4.2, calculating the characteristic vector of each characteristic point and the characteristic value of the neighborhood covariance matrix, wherein the characteristic vector and the characteristic value are as follows:
COV·v j ＝λ j ·v j
where j ∈ {1,2,3}, λ j Is an eigenvalue, v, of a covariance matrix j Is the relevant feature vector.
Calculating a Key Point K i Curvature and neighborhood space Surface ir The interior point curvatures H are summed as follows:
wherein λ is 1 Is λ j The smallest eigenvalue of (d).
Step 4.3: calculating a Key Point K i And neighborhood space Surface ik The sum of the distance values of the inner points is as follows:
wherein, K ik Is K i Points in the neighborhood of the radius, i is the index of the keypoint, and k is the index of the point in the neighborhood.
Step 4.3: and (4) constructing a descriptor of the feature histogram of the neighborhood points according to the three values in the steps 4.1, 4.2 and 4.3.
The step 5 specifically comprises the following steps:
step 5.1: first, for a source key point K in a new 3D space iRFsource Find the distance threshold T d Inner most recent target keypoint list K iRFtarget ；
Step 5.2: and finding out an accurate corresponding point by calculating the Euclidean distance in a descriptor of the 'neighborhood point feature histogram' to generate a matching point pair C.
Step 5.3: for corresponding characteristic point pairs, a three-dimensional rigid transformation matrix T from the source point cloud P to the target point cloud Q is obtained by using a singular value decomposition algorithm i ；
Step 5.4: from an estimated three-dimensional rigid transformation matrix T i Respectively carrying out coordinate conversion on each pair of corresponding characteristic points, judging whether each pair of corresponding characteristic points belong to interior points according to Euclidean distances between the corresponding characteristic points after the coordinate conversion, and obtaining the number of the interior points under the estimated three-dimensional rigid transformation matrix, wherein the expression is described as follows;
wherein k is 1,2, …, N c ,N c And E, representing the logarithm of the corresponding characteristic points in the characteristic matching point pair C, wherein epsilon is a distance threshold value.
Step 5.5: repeating the steps 5.4 and 5.5 to obtain the number of inner points under each pair of corresponding feature point estimated three-dimensional rigid transformation matrix, and determining the estimated three-dimensional rigid transformation matrix with the largest number of the inner points as an initial three-dimensional rigid transformation matrix T, which is described as follows:
and registering the source point cloud to the target point cloud by the initial transformation matrix T to obtain the initial position from the source point cloud to the target point cloud.
Compared with the prior art, the invention has the following advantages:
(1) the present algorithm greatly reduces the matching point pair search space by using the key point coordinate locations in the new 3D space.
(2) The method is easy to combine with various advanced key point detectors;
(3) and a 'neighborhood point feature histogram' descriptor is used for further determining matching point pairs, so that the calculation dimensionality of local features is reduced, the calculation speed is greatly increased, and the registration efficiency is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principles of the invention. In the drawings:
fig. 1 is a flow chart of a point cloud registration algorithm designed by the invention.
FIG. 2 is a schematic diagram of the local feature sub-calculation of the low-dimensional point cloud.
Fig. 3 is a schematic diagram of "neighborhood point feature histogram" descriptor calculation.
FIG. 4 is an initial pose graph of a source point cloud and a target point cloud.
Fig. 5 shows the registration result after the RANSAC random sampling consistency algorithm.
Detailed Description
The following further describes the embodiments of the present invention with reference to the drawings. Under a Windows 10 operating system, Visual Studio2019 is selected as a programming tool to perform registration processing on multi-view point cloud data acquired by the three-dimensional measuring equipment. In the example, a 'deep-view intelligent' SR7080 scanner is adopted to obtain camera module point cloud data, and finally, a relatively accurate registration result is obtained. The drawings are for illustrative purposes only and are not to be construed as limiting the patent;
the figure is an algorithm flow chart of the whole process of the invention.
The point clouds with different visual angles need to be registered in the post-processing of the three-dimensional point cloud, and the method aims at the problems of low feature identification degree, large descriptor dimension, long registration time and the like in the existing automatic registration algorithm based on feature extraction and description. The invention provides a point cloud registration algorithm based on a point cloud local feature descriptor. The point cloud is transferred to a new 3D space by constructing a local reference coordinate system, key points of similar curved surfaces are close to each other, and a matching pair search space is greatly reduced. And meanwhile, the matching point pairs are further constrained and sought by utilizing the normal vector, the curvature and the distance information. And then obtaining an accurate corresponding point pair by using a random sampling consistency algorithm (RANSAC), calculating initial registration parameters by using a singular value decomposition method, and finally accurately registering point clouds by using an iterative closest point method (ICP).
The method comprises the following specific implementation steps:
step 1: two groups of data of a source point cloud P and a target point cloud Q are collected, the point clouds are required to be acquired without visual angles and have certain overlapping degree, and then a point cloud normal vector is calculated. FIG. 4 is a point cloud initial pose map.
Step 2: downsampling using voxel grid centroid neighbors. The method specifically comprises the following steps:
(1) obtaining a set of keypoint points P using a voxel filter i 、Q i 。
(2) Traversing the point cloud P, Q by using KD-tree to find and combine the point set P i 、Q i The nearest neighbor point of each point in the key point set P s 、Q t 。
And step 3: and constructing a 3D descriptor to obtain a list to be matched.
Firstly, downsampling source point cloud key points and target point cloud key points and local points through a constructed local reference coordinate systemThe keypoints are converted into the new 3D space by multiplication with reference to the coordinate system to constitute the "3D" descriptor. In the new 3D space, each source point cloud key point is subjected to the process of taking the radius as T d And obtaining a list to be matched by radial searching.
And 4, step 4: and constructing a descriptor of the feature histogram of the neighborhood points and determining matching point pairs.
Calculating the average distance, curvature and change and normal angle sum of the source point cloud key point and the target point cloud key point and the corresponding neighborhood points in the radius neighborhood, encoding the last three dimensions of the six-dimensional feature descriptor, and finally obtaining the neighborhood point feature histogram descriptor. Fig. 3 is a schematic diagram of "neighborhood point feature histogram" descriptor calculation. In the list to be matched, the exact corresponding neighbor of the source key point is found in the descriptor of the 'neighborhood point feature histogram' of the rear three-dimensional point through Euclidean measurement, and the exact match of the source key point under consideration is found from the list to be matched generated above.
And 5: after a random sampling consistency algorithm is used, mismatching point pairs are removed, and the attached figure 5 shows a point cloud registration result.
Through analysis, the point cloud registration algorithm designed by the algorithm provides a six-dimensional local feature descriptor, and the descriptor has the characteristics of low dimensionality, strong descriptiveness, high extraction speed and the like, obviously improves the point cloud registration speed, and has excellent registration precision.
The above embodiments are provided to further explain the objects, technical solutions and advantages of the present invention in detail, it should be understood that the above embodiments are merely exemplary embodiments of the present invention and are not intended to limit the scope of the present invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (5)
1. A point cloud registration method based on a low-dimensional point cloud local feature descriptor is characterized by comprising the following steps: the method comprises the following steps:
step 1: acquiring point cloud data and calculating a normal vector of the point cloud data;
and 2, step: selecting a voxel gravity center nearest neighbor point as a key point through point cloud down-sampling;
and step 3: the method comprises the steps of obtaining a neighborhood surface point set by key points, then calculating a correction covariance matrix of point cloud, transferring the key points to a new 3D space by establishing a local reference coordinate system, and enabling new 3D coordinates of the key points to form a 3D descriptor.
And 4, step 4: and coding the neighborhood characteristics of the point cloud in the next 3 dimensions by calculating the sum of the included angles, curvatures and distances of normal vectors of the point cloud and providing a descriptor of a neighborhood point characteristic histogram. The '3D' descriptor and the 'neighborhood point feature histogram' descriptor jointly form a local feature descriptor of the low-dimensional point cloud.
And 5: matching is carried out through the geometric surface description of the key points, and a rotation matrix and a translation matrix are calculated by utilizing a random sample consensus (RANSAC) algorithm to complete point cloud registration.
2. The point cloud local feature descriptor-based point cloud registration method of claim 1, wherein: the step 2 adopts the following steps to extract key points:
step 2.1: establishing KD-tree for point cloud, counting total number N of points in point cloud, calculating minimum bounding box and volume V of point cloud, and setting L x 、L y 、L z The length of the bounding box in the x, y, z directions, respectively. The KD-Tree is named as K-Dimensional Tree, and a KNN retrieval data structure is used for storing a K-Dimensional space vector point set and performing quick retrieval.
Step 2.2: setting the side length of the small cube to
The point cloud is divided into small cubes of n ═ l × w × h. Wherein: s is a scale factor, alpha is a scale factor for adjusting the side length according to the number of the point clouds, and L is ceil (L) x /L)，w＝ceil(L y /L)，h＝ceil(L z /L), ceil is an rounding-up function.
Step 2.3: calculating the gravity center of each non-empty small cube as p 0 (x 0 ，y 0 ，z 0 ) Build a new set of points { P } c And c is 1,2,3, …, n, wherein
Step 2.4: traversing the point cloud P, Q by using kd-tree to find a point set P i 、Q i The nearest neighbor point of each point in the key point set P s 、Q t 。
3. The point cloud local feature descriptor-based point cloud registration method of claim 1, wherein: said step 3 encodes the 3D keypoint locations using the following steps to construct a "3D" descriptor.
Step 3.1: calculating a Key Point K i Surface radius r neighborhood space Surface ik Internal centroid coordinate mean pt ；
Step 3.2: surface of ik Coordinates of all points in the middle minus the centroid coordinate mean pt Effectively generate a new surface
Step 3.3: let the key point Ki subtract the centroid coordinate mean pt To obtain
Step 3.4: we draw from a new surface
In estimating a local reference coordinate system [ RF ]] 3×3 To point out the key
Converting to a new 3D space by the following conversion process;
wherein [ K ] iRF ] 3×1 Is the generated "3D" descriptor.
4. The point cloud local feature descriptor-based point cloud registration method of claim 1, wherein: the step 4 adopts the following steps to calculate the feature histogram of the neighborhood point:
step 4.1: calculating a Key Point K i The sum of cosine values of an included angle between the normal vector and the normal vector of the surface neighborhood point;
step 4.2: calculating a Key Point K i The sum of the curvature of (a) and the curvature value of the surface neighborhood point;
step 4.3: calculating a Key Point K i And Surface ik The sum of the distances between the points;
step 4.4: and (4) combining the three characteristic values in the steps 4.1, 4.2 and 4.3 to form a neighborhood point characteristic histogram.
5. The point cloud local feature descriptor-based point cloud registration method of claim 1, wherein: the step 5 adopts the following steps to match and calculate a rigid transformation matrix:
step 5.1: first, for a source key point K in a new 3D space iRFsource Radial search threshold T d Inner most recent target keypoint list K iRFtarget ；
Step 5.2: we find the exact corresponding point by using Euclidean distance through the descriptor of 'neighborhood point feature histogram', and generate the matching point pair.
Step 5.3: for the corresponding characteristic point pairs, obtaining a three-dimensional rigid transformation matrix from the source point cloud to the target point cloud by using a singular value decomposition algorithm;
step 5.4: according to the estimated three-dimensional rigid transformation matrix, performing coordinate transformation on each pair of corresponding characteristic points respectively, and obtaining the number of interior points under the three-dimensional rigid transformation matrix according to the Euclidean distance between the corresponding characteristic points after the coordinate transformation;
step 5.5: and (5) repeating the steps 5.4 and 5.5, obtaining the number of the inner points under each pair of corresponding characteristic points and estimating the three-dimensional rigid transformation matrix, determining the estimated three-dimensional rigid transformation matrix with the largest number of the inner points as the final three-dimensional rigid transformation matrix, and finishing point cloud registration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210611042.3A CN114972459B (en) | 2022-05-31 | 2022-05-31 | Point cloud registration method based on low-dimensional point cloud local feature descriptor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210611042.3A CN114972459B (en) | 2022-05-31 | 2022-05-31 | Point cloud registration method based on low-dimensional point cloud local feature descriptor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114972459A true CN114972459A (en) | 2022-08-30 |
| CN114972459B CN114972459B (en) | 2023-09-08 |
Family
ID=82958184
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210611042.3A Active CN114972459B (en) | 2022-05-31 | 2022-05-31 | Point cloud registration method based on low-dimensional point cloud local feature descriptor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114972459B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115830340A (en) * | 2022-11-16 | 2023-03-21 | 白城师范学院 | Point cloud target identification method and device and storage medium |
| CN115953604A (en) * | 2023-03-13 | 2023-04-11 | 泰安市金土地测绘整理有限公司 | Real estate geographic information mapping data acquisition method |
| CN116046017A (en) * | 2023-01-18 | 2023-05-02 | 上汽通用五菱汽车股份有限公司 | Calibration method and device for measuring path, storage medium and computer equipment |
| CN116434221A (en) * | 2023-06-13 | 2023-07-14 | 中科航迈数控软件(深圳)有限公司 | Workpiece shape recognition method, device, terminal equipment and computer storage medium |
| CN116817771A (en) * | 2023-08-28 | 2023-09-29 | 南京航空航天大学 | Aerospace part coating thickness measurement method based on cylindrical voxel characteristics |
| CN117237428A (en) * | 2023-11-15 | 2023-12-15 | 南京航空航天大学 | Data registration method, device and medium for three-dimensional point cloud |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109215129A (en) * | 2017-07-05 | 2019-01-15 | 中国科学院沈阳自动化研究所 | A kind of method for describing local characteristic based on three-dimensional point cloud |
| CN110335297A (en) * | 2019-06-21 | 2019-10-15 | 华中科技大学 | A kind of point cloud registration method based on feature extraction |
| CN110490912A (en) * | 2019-07-17 | 2019-11-22 | 哈尔滨工程大学 | 3D-RGB point cloud registration method based on local gray level sequence model descriptor |
| CN111080684A (en) * | 2019-12-12 | 2020-04-28 | 哈尔滨工程大学 | Point cloud registration method for point neighborhood scale difference description |
| US20200402244A1 (en) * | 2019-06-19 | 2020-12-24 | Faro Technologies, Inc. | Method and apparatus for registering three-dimensional point clouds |
| CN112651408A (en) * | 2021-01-07 | 2021-04-13 | 华中科技大学 | Point-to-point transformation characteristic-based three-dimensional local surface description method and system |
| CN114332172A (en) * | 2021-10-14 | 2022-04-12 | 中北大学 | Improved laser point cloud registration method based on covariance matrix |
-
2022
- 2022-05-31 CN CN202210611042.3A patent/CN114972459B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109215129A (en) * | 2017-07-05 | 2019-01-15 | 中国科学院沈阳自动化研究所 | A kind of method for describing local characteristic based on three-dimensional point cloud |
| US20200402244A1 (en) * | 2019-06-19 | 2020-12-24 | Faro Technologies, Inc. | Method and apparatus for registering three-dimensional point clouds |
| CN110335297A (en) * | 2019-06-21 | 2019-10-15 | 华中科技大学 | A kind of point cloud registration method based on feature extraction |
| CN110490912A (en) * | 2019-07-17 | 2019-11-22 | 哈尔滨工程大学 | 3D-RGB point cloud registration method based on local gray level sequence model descriptor |
| CN111080684A (en) * | 2019-12-12 | 2020-04-28 | 哈尔滨工程大学 | Point cloud registration method for point neighborhood scale difference description |
| CN112651408A (en) * | 2021-01-07 | 2021-04-13 | 华中科技大学 | Point-to-point transformation characteristic-based three-dimensional local surface description method and system |
| CN114332172A (en) * | 2021-10-14 | 2022-04-12 | 中北大学 | Improved laser point cloud registration method based on covariance matrix |
Non-Patent Citations (3)
| Title |
|---|
| YAO X: ""cloud registration algorithm based on curvature feature similarity"", 《MEASUREMENT》, pages 1 - 12 * |
| 张彬: ""基于体素下采样和关键点提取的点云自动配准"", 《激光与光电子学进展》, vol. 57, no. 4, pages 1 - 9 * |
| 李新春: ""基于邻域特征点提取和匹配的点云配准"", 《光子学报》, vol. 49, no. 4, pages 1 - 11 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115830340A (en) * | 2022-11-16 | 2023-03-21 | 白城师范学院 | Point cloud target identification method and device and storage medium |
| CN115830340B (en) * | 2022-11-16 | 2023-11-21 | 白城师范学院 | Point cloud target identification method and device and storage medium |
| CN116046017A (en) * | 2023-01-18 | 2023-05-02 | 上汽通用五菱汽车股份有限公司 | Calibration method and device for measuring path, storage medium and computer equipment |
| CN116046017B (en) * | 2023-01-18 | 2024-05-07 | 上汽通用五菱汽车股份有限公司 | Calibration method and device for measuring path, storage medium and computer equipment |
| CN115953604A (en) * | 2023-03-13 | 2023-04-11 | 泰安市金土地测绘整理有限公司 | Real estate geographic information mapping data acquisition method |
| CN115953604B (en) * | 2023-03-13 | 2023-05-30 | 泰安市金土地测绘整理有限公司 | Real estate geographic information mapping data acquisition method |
| CN116434221A (en) * | 2023-06-13 | 2023-07-14 | 中科航迈数控软件(深圳)有限公司 | Workpiece shape recognition method, device, terminal equipment and computer storage medium |
| CN116434221B (en) * | 2023-06-13 | 2023-09-01 | 中科航迈数控软件(深圳)有限公司 | Workpiece shape recognition method, device, terminal equipment and computer storage medium |
| CN116817771A (en) * | 2023-08-28 | 2023-09-29 | 南京航空航天大学 | Aerospace part coating thickness measurement method based on cylindrical voxel characteristics |
| CN116817771B (en) * | 2023-08-28 | 2023-11-17 | 南京航空航天大学 | Aerospace part coating thickness measurement method based on cylindrical voxel characteristics |
| CN117237428A (en) * | 2023-11-15 | 2023-12-15 | 南京航空航天大学 | Data registration method, device and medium for three-dimensional point cloud |
| CN117237428B (en) * | 2023-11-15 | 2024-02-20 | 南京航空航天大学 | Data registration method, device and medium for three-dimensional point cloud |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114972459B (en) | 2023-09-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114972459B (en) | Point cloud registration method based on low-dimensional point cloud local feature descriptor | |
| CN109887015B (en) | Point cloud automatic registration method based on local curved surface feature histogram | |
| CN111080684B (en) | Point cloud registration method for point neighborhood scale difference description | |
| CN103729643B (en) | The identification of three dimensional object in multi-mode scene and posture are determined | |
| CN107833181B (en) | Three-dimensional panoramic image generation method based on zoom stereo vision | |
| CN102236794B (en) | Recognition and pose determination of 3D objects in 3D scenes | |
| CN111553409B (en) | Point cloud identification method based on voxel shape descriptor | |
| CN114677418B (en) | Registration method based on point cloud feature point extraction | |
| CN109523582B (en) | Point cloud coarse registration method considering normal vector and multi-scale sparse features | |
| CN104090972A (en) | Image feature extraction and similarity measurement method used for three-dimensional city model retrieval | |
| CN110930456A (en) | Three-dimensional identification and positioning method of sheet metal part based on PCL point cloud library | |
| CN110111375B (en) | Image matching gross error elimination method and device under Delaunay triangulation network constraint | |
| CN108830888B (en) | Coarse matching method based on improved multi-scale covariance matrix characteristic descriptor | |
| CN113223173B (en) | Three-dimensional model reconstruction migration method and system based on graph model | |
| CN111242221A (en) | Image matching method, system and storage medium based on image matching | |
| CN108182705A (en) | A kind of three-dimensional coordinate localization method based on machine vision | |
| Yi et al. | Survey of structure from motion | |
| CN115861397A (en) | Point cloud registration method based on improved FPFH-ICP | |
| CN112396641A (en) | Point cloud global registration method based on congruent two-baseline matching | |
| Lu et al. | Point cloud registration algorithm fusing of super 4pcs and icp based on the key points | |
| CN110246165B (en) | Method and system for improving registration speed of visible light image and SAR image | |
| CN114463396B (en) | Point cloud registration method utilizing plane shape and topological graph voting | |
| Lu et al. | Automatic point cloud registration algorithm based on the feature histogram of local surface | |
| CN115082716A (en) | Multi-source point cloud rough matching algorithm for road fine reconstruction | |
| CN112785631A (en) | Point cloud data registration method based on DLG |
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 |