CN111986308B - Point cloud normal and curvature change based double-constraint surface error salient point identification method - Google Patents
Point cloud normal and curvature change based double-constraint surface error salient point identification method Download PDFInfo
- Publication number
- CN111986308B CN111986308B CN202010675830.XA CN202010675830A CN111986308B CN 111986308 B CN111986308 B CN 111986308B CN 202010675830 A CN202010675830 A CN 202010675830A CN 111986308 B CN111986308 B CN 111986308B
- Authority
- CN
- China
- Prior art keywords
- point
- point cloud
- error
- curvature
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/005—General purpose rendering architectures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- 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
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20101—Interactive definition of point of interest, landmark or seed
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Graphics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Length Measuring Devices With Unspecified Measuring Means (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a point cloud normal and curvature change double-constraint-based surface error salient point identification method, which comprises the following steps: scanning the workpiece by using a scanner to obtain point cloud data; filtering and denoising the point cloud data, and dividing the point cloud by adopting a kd-tree method; establishing a neighborhood relation between each point and adjacent points in the kd-tree by using K neighbor difference finding, and solving a point cloud normal and curvature; traversing the whole point cloud by adopting a region growing algorithm based on depth-first search, and adding double constraints of point cloud normal and curvature change as limiting conditions of region growing; screening the area obtained by the area growth, and removing the minimum and maximum areas; merging the remaining adjacent areas to obtain an identified error significant point cloud; and carrying out boundary extraction on the identified error salient point cloud, and determining the position information of the error salient point. The beneficial effects provided by the invention are as follows: the searching efficiency, the accuracy and the position accuracy of the point cloud error significant point identification are improved.
Description
Technical Field
The invention relates to the field of three-dimensional reconstruction, in particular to a point cloud normal and curvature change-based double-constraint surface error salient point identification method.
Background
The popularization of digital equipment and 3D scanners ensures that the acquisition of the point cloud data is quicker and more accurate. Compared with image segmentation and grid segmentation, the point cloud data keeps the subtle characteristics of the three-dimensional model and is convenient to store, so that the point cloud processing of the three-dimensional model has become a research hot spot.
The point cloud segmentation divides the point cloud according to the characteristics of space, geometry, texture and the like, so that the point clouds in the same division have similar characteristics. The effective segmentation of the point cloud is a precondition that better hole repair and curved surface reconstruction can be performed.
The identification of the salient points of the curved surface errors is a typical point cloud segmentation problem, but the traditional point cloud segmentation algorithm is low in calculation efficiency when the salient points of the errors are identified, and the identification precision is poor, and particularly the salient points of the errors on the curved surface are not easy to identify.
Disclosure of Invention
In view of the above, the invention provides a method for identifying points with obvious errors of a double-constraint curved surface based on point cloud normal and curvature change, which comprises the following steps:
s101: scanning the processed workpiece by using a three-dimensional scanner to obtain point cloud data;
s102: filtering and denoising the point cloud data to obtain filtered and denoised point cloud data;
s103: dividing the point cloud data subjected to filtering denoising by adopting a kd-tree method, and solving the normal line and curvature of the point cloud data;
s104, performing S104; traversing the filtered and denoised point cloud data by adopting a region growing algorithm based on depth-first search, and adding double constraints of point cloud normals and curvature changes as limiting conditions of region growing;
s105: screening the region obtained by the region growth, and removing the minimum and maximum regions to obtain a screened region;
s106: merging the screened areas to obtain an identified error significant point cloud;
s107: and carrying out boundary extraction on the identified error salient point cloud to obtain the position information of the error salient point.
Further, in step S101, the point cloud data includes x, y, z three-dimensional features and regions of error salient points to be identified.
Further, in step S102, filtering and denoising the point cloud data to obtain filtered and denoised point cloud data, specifically: and setting the voxel grid size of the point cloud data, carrying out voxel filtering on the point cloud data, reducing the scale of the point cloud, and obtaining the filtered and denoised point cloud data.
Further, step S103 specifically includes: establishing a neighborhood relation between each point and adjacent points thereof by using a K neighbor method in a kd-tree, and setting the number K of the neighborhood points 1 Extracting k adjacent to each point in the filtered and denoised point cloud data 1 Sub-point cloud composed of individual points and based on this k 1 And the normal line and the curvature of the point are obtained by the sub-point cloud formed by the points.
Further, step S104 specifically includes:
s201: performing ascending order sorting on the filtered and denoised point cloud data according to the respective curvature sizes from small to large;
s202: creating a container with the same size as the filtered and denoised point cloud data, wherein all values in the container are preset to be 0 to indicate that the container is not accessed;
s203: selecting a point with the minimum curvature in points which are not accessed at present as an initial seed point M, pressing the initial point M into a queue, and simultaneously assigning a container corresponding to the seed point M as 1 to indicate that the point is accessed;
s204: according to the seed point M, k is obtained according to step S103 1 The neighborhood points are selected, and the seed points M are popped up from the queue;
s205: k for seed point M 1 Traversing the neighborhood points in sequence, and adding point cloud normal threshold constraint and point cloud curvature change threshold constraint; if the seed point M is k 1 Setting the container corresponding to a certain point in the neighborhood points and the normal angle change of the seed point M to be 1 if the normal angle change of the certain point and the seed point M is smaller than the normal threshold value and the curvature change of the certain point and the seed point M is also smaller than the curvature threshold value, and pressing the container into a queue to obtain an updated queue, otherwise, keeping the container corresponding to the certain point to be 0;
s206: step S204, after traversing, selecting a point M of the head of the updated queue 1 As a new seed point, repeating the steps S204-S205, and judging whether the queue is empty, if yes, proceeding to step S207, otherwise continuing to repeat the steps S204-S205;
S207:the queue is empty and represents the area generated by the point M with the smallest curvature at this time 1 Finishing the generation; at this time, returning to step S203, selecting another point with the smallest curvature among the points which are not accessed currently as a new seed point, repeating steps S203 to S207 to generate an area of the point, and continuing to repeat steps S203 to S207 until all the generation areas corresponding to the points with the smallest curvature which are not accessed are generated.
Further, in step S105, the region obtained by the region growth is screened, and the minimum and maximum regions are removed, so as to obtain a screened region, where the screening conditions include, but are not limited to: the number and the position of the point clouds; the maximum area is specifically a point cloud area without significant points of error; the minimum area is specifically a point cloud area in which noise points cannot be filtered in step S102.
Further, in step S106, the screened areas are combined to obtain the identified point cloud with significant error, which specifically includes: combining according to the adjacency relation of the screened areas, combining the screened areas with adjacency relation into a point cloud which is used as the point cloud of the same significant error point, and not combining the screened areas without adjacency relation to be used as the point cloud of different significant error points.
In S107, boundary extraction is performed on the identified point cloud of the error significant point, so as to obtain position information of the error significant point, which specifically includes: setting a boundary K neighbor search parameter K2 according to the error significant point cloud normal and the kd-tree dividing result to obtain a boundary point cloud of an error significant point; calculating the center of gravity P of the error salient point by using a boundary point cloud of the error salient point and the length and width a and b of the minimum bounding rectangle; the position information of the error significant point is the center of gravity P and the length and width a and b of the minimum bounding rectangle.
The beneficial effects provided by the invention are as follows: the searching efficiency, the accuracy and the position accuracy of the point cloud error significant point identification are improved.
Drawings
FIG. 1 is a flow diagram of a method for identifying points of significance of a double-constraint surface error based on point cloud normals and curvature changes;
FIG. 2 is an implementation circle as a theoretical boundary;
FIG. 3 is a graph of error salient point recognition effect according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be further described with reference to the accompanying drawings.
Referring to fig. 1, an embodiment of the present invention provides a method for identifying points of interest of a double-constraint surface error based on a point cloud normal and curvature change, including the following steps:
s101: scanning the processed workpiece by using a three-dimensional scanner to obtain point cloud data;
the workpiece tested in this example was a 3d print with protrusions on one plane.
S102: filtering and denoising the point cloud data to obtain filtered and denoised point cloud data;
s103: dividing the point cloud data subjected to filtering denoising by adopting a kd-tree method, and solving the normal line and curvature of the point cloud data; firstly removing NAN points from the obtained point cloud, and then reducing the scale of the point cloud from 140w to 60w by voxel filtering; in this embodiment, the kd-Tree method is performed based on a relevant library function of a PCL point cloud library.
S104, performing S104; traversing the filtered and denoised point cloud data by adopting a region growing algorithm based on depth-first search, and adding double constraints of point cloud normals and curvature changes as limiting conditions of region growing;
s105: screening the region obtained by the region growth, and removing the minimum and maximum regions to obtain a screened region; in the embodiment, the minimum and maximum areas are removed by limiting the number of the point clouds. Setting the lower limit of the number of the point clouds as 100 and the upper limit as 10w, eliminating the point clouds outside the area, and setting the rest area as the area where the significant error points are located
S106: merging the screened areas to obtain an identified error significant point cloud; because the original point cloud is subjected to kd-tree division, the neighbor relation of each point in the point cloud is still reserved in the vector container, whether the areas are adjacent or not is rapidly judged by utilizing the neighbor relation, and the adjacent areas are combined into the same area. After this step, each error significance point will correspond to only one region.
S107: and carrying out boundary extraction on the identified error salient point cloud to obtain the position information of the error salient point.
Further, in step S101, the point cloud data includes x, y, z three-dimensional features and regions of error salient points to be identified.
Further, in step S102, filtering and denoising the point cloud data to obtain filtered and denoised point cloud data, specifically: and setting the voxel grid size of the point cloud data, carrying out voxel filtering on the point cloud data, reducing the scale of the point cloud, and obtaining the filtered and denoised point cloud data.
The step S103 specifically includes: establishing a neighborhood relation between each point and adjacent points thereof by using a K neighbor method in a kd-tree, and setting the number K of the neighborhood points 1 Extracting k adjacent to each point in the filtered and denoised point cloud data 1 Sub-point cloud composed of individual points and based on this k 1 And the normal line and the curvature of the point are obtained by the sub-point cloud formed by the points.
In this embodiment, the K-nearest neighbor difference finding number is set to 10, and K-nearest neighbor difference finding is performed. A vector container is established to store the adjacent points after K neighbor difference finding of each point. After 10 neighboring point clouds of the selected point are extracted, surface normal estimation and curvature calculation are performed using these point clouds.
The step S104 specifically includes:
s201: performing ascending order sorting on the filtered and denoised point cloud data according to the respective curvature sizes from small to large;
s202: creating a container with the same size as the filtered and denoised point cloud data, wherein all values in the container are preset to be 0 to indicate that the container is not accessed;
s203: selecting a point with the minimum curvature in points which are not accessed at present as an initial seed point M, pressing the initial point M into a queue, and simultaneously assigning a container corresponding to the seed point M as 1 to indicate that the point is accessed;
s204: according to the seed point M, k is obtained according to step S103 1 The neighborhood points are selected, and the seed points M are popped up from the queue;
s205: k for seed point M 1 Traversing the neighborhood points in sequence, and adding point cloud normal threshold constraint and point cloud curvature change threshold constraint; if the seed point M is k 1 Setting the container corresponding to a certain point in the neighborhood points and the normal angle change of the seed point M to be 1 if the normal angle change of the certain point and the seed point M is smaller than the normal threshold value and the curvature change of the certain point and the seed point M is also smaller than the curvature threshold value, and pressing the container into a queue to obtain an updated queue, otherwise, keeping the container corresponding to the certain point to be 0; in this embodiment, the normal angle threshold is set to pi/20 and the curvature change threshold is set to 0.04.
S206: step S204, after traversing, selecting a point M of the head of the updated queue 1 As a new seed point, repeating the steps S204-S205, and judging whether the queue is empty, if yes, proceeding to step S207, otherwise continuing to repeat the steps S204-S205;
s207: the queue is empty and represents the area generated by the point M with the smallest curvature at this time 1 Finishing the generation; at this time, returning to step S203, selecting another point with the smallest curvature among the points which are not accessed currently as a new seed point, repeating steps S203 to S207 to generate an area of the point, and continuing to repeat steps S203 to S207 until all the generation areas corresponding to the points with the smallest curvature which are not accessed are generated.
In step S104, a double constraint of the point cloud normal and curvature change is added as a constraint condition for region growth, in order to make the boundary determination of the significant error point more accurate. Referring to fig. 2, the implementation circle shown in fig. 2 is a theoretical boundary, and when only normal constraint is adopted, normal variation is not obvious when there is a transition from a straight line to an arc, which results in a larger difference between the boundary adopting only normal constraint and the actual boundary. However, in the case of transition from a straight line to an arc, the normal line does not change significantly but the curvature is abrupt at the transition, so that if the two are combined, the regions can be divided more precisely to determine the boundary.
Screening the region obtained by the region growth in step S105, and removing the minimum and maximum regions to obtain a screened region, wherein the screening conditions include, but are not limited to: the number and the position of the point clouds; the maximum area is specifically a point cloud area without significant points of error; the minimum area is specifically a point cloud area in which noise points cannot be filtered in step S102.
Since the error salient points (such as the welding spots, pits, protrusions, etc.) are irregular in step S106, the same error salient point may be divided into a plurality of areas under the above-mentioned area growing algorithm, but the plurality of areas have an adjacent relationship, and different error salient points do not have an adjacent relationship, otherwise, it may be considered that the 2 error salient points may be combined into one larger error salient point. In step S106, the screened areas are combined to obtain the identified point cloud with significant error, which specifically includes: combining according to the adjacency relation of the screened areas, combining the screened areas with adjacency relation into a point cloud which is used as the point cloud of the same significant error point, and not combining the screened areas without adjacency relation to be used as the point cloud of different significant error points.
In S107, boundary extraction is performed on the identified point cloud of the error significant point, so as to obtain position information of the error significant point, which specifically includes: setting a boundary K neighbor search parameter K2 according to the error significant point cloud normal and the kd-tree dividing result to obtain a boundary point cloud of an error significant point; calculating the center of gravity P of the error salient point by using a boundary point cloud of the error salient point and the length and width a and b of the minimum bounding rectangle; the position information of the error significant point is the center of gravity P and the length and width a and b of the minimum bounding rectangle.
Referring to fig. 3, fig. 3 is an effect diagram of an error salient point cloud obtained in the embodiment of the invention; the white area in fig. 3 is the finally detected error salient point cloud.
In general, the above technical solutions conceived by the present invention, compared with the prior art, enable the following beneficial effects to be obtained:
(1) Aiming at the traditional error significant point identification of the processed workpiece, the invention provides an automatic error significant point identification mode in a point cloud form. The method fully utilizes the normal and curvature changes of the point cloud, can quickly and accurately identify, and is simpler and quicker than the traditional measuring tool measuring method.
(2) The depth-first search is adopted to carry out point cloud traversal, so that the search efficiency is effectively improved, and the quick identification of the error significant points is realized.
(3) The double constraint conditions of point cloud normal variation and curvature variation are adopted, so that the identification accuracy and the position accuracy of the point cloud error significant points are higher.
The invention has the beneficial effects that: the searching efficiency, the accuracy and the position accuracy of the point cloud error significant point identification are improved.
The above-described embodiments of the invention and features of the embodiments may be combined with each other without conflict.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.
Claims (8)
1. The method for identifying the point cloud normal and curvature change-based double-constraint surface error salient point is characterized by comprising the following steps of:
s101: scanning the processed workpiece by using a three-dimensional scanner to obtain point cloud data;
s102: filtering and denoising the point cloud data to obtain filtered and denoised point cloud data;
s103: dividing the point cloud data subjected to filtering denoising by adopting a kd-tree method, and solving the normal line and curvature of the point cloud data;
s104: traversing the filtered and denoised point cloud data by adopting a region growing algorithm based on depth-first search, and adding double constraints of point cloud normals and curvature changes as limiting conditions of region growing;
s105: screening the region obtained by the region growth, and removing the minimum and maximum regions to obtain a screened region;
s106: merging the screened areas to obtain an identified error significant point cloud;
s107: and carrying out boundary extraction on the identified error salient point cloud to obtain the position information of the error salient point.
2. The method for identifying the point cloud normal and curvature change-based double-constraint surface error salient point according to claim 1, which is characterized by comprising the following steps: in step S101, the point cloud data includes x, y, z three-dimensional features and regions of error salient points to be identified.
3. The method for identifying the point cloud normal and curvature change-based double-constraint surface error salient point according to claim 1, which is characterized by comprising the following steps: in step S102, filtering and denoising the point cloud data to obtain filtered and denoised point cloud data, which specifically includes: and setting the voxel grid size of the point cloud data, carrying out voxel filtering on the point cloud data, reducing the scale of the point cloud, and obtaining the filtered and denoised point cloud data.
4. The method for identifying the point cloud normal and curvature change-based double-constraint surface error salient point according to claim 1, which is characterized by comprising the following steps: the step S103 specifically includes: establishing a neighborhood relation between each point and adjacent points thereof by using a K neighbor method in a kd-tree, and setting the number K of the neighborhood points 1 Extracting k adjacent to each point in the filtered and denoised point cloud data 1 Sub-point cloud composed of individual points and based on this k 1 And the normal line and the curvature of the point are obtained by the sub-point cloud formed by the points.
5. The method for identifying the point cloud normal and curvature change-based double-constraint surface error salient point according to claim 4, which is characterized by comprising the following steps: the step S104 specifically includes:
s201: performing ascending order sorting on the filtered and denoised point cloud data according to the respective curvature sizes from small to large;
s202: creating a container with the same size as the filtered and denoised point cloud data, wherein all values in the container are preset to be 0 to indicate that the container is not accessed;
s203: selecting a point with the minimum curvature in points which are not accessed at present as an initial seed point M, pressing the initial point M into a queue, and simultaneously assigning a container corresponding to the seed point M as 1 to indicate that the point is accessed;
s204: according to the seed point M, k is obtained according to step S103 1 The neighborhood points are selected, and the seed points M are popped up from the queue;
s205: k for seed point M 1 Traversing the neighborhood points in sequence, and adding point cloud normal threshold constraint and point cloud curvature change threshold constraint; if the seed point M is k 1 Setting the container corresponding to a certain point in the neighborhood points and the normal angle change of the seed point M to be 1 if the normal angle change of the certain point and the seed point M is smaller than the normal threshold value and the curvature change of the certain point and the seed point M is also smaller than the curvature threshold value, and pressing the container into a queue to obtain an updated queue, otherwise, keeping the container corresponding to the certain point to be 0;
s206: step S204, after traversing, selecting a point M of the head of the updated queue 1 As a new seed point, repeating the steps S204-S205, and judging whether the queue is empty, if yes, proceeding to step S207, otherwise continuing to repeat the steps S204-S205;
s207: the queue is empty and represents the area generated by the point M with the smallest curvature at this time 1 Finishing the generation; at this time, returning to step S203, selecting another point with the smallest curvature among the points which are not accessed currently as a new seed point, repeating steps S203 to S207 to generate an area of the point, and continuing to repeat steps S203 to S207 until all the generation areas corresponding to the points with the smallest curvature which are not accessed are generated.
6. The method for identifying the point cloud normal and curvature change-based double-constraint surface error salient point according to claim 1, which is characterized by comprising the following steps: screening the region obtained by the region growth in step S105, and removing the minimum and maximum regions to obtain a screened region, wherein the screening conditions include, but are not limited to: the number and the position of the point clouds; the maximum area is specifically a point cloud area without significant points of error; the minimum area is specifically a point cloud area in which noise points cannot be filtered in step S102.
7. The method for identifying the point cloud normal and curvature change-based double-constraint surface error salient point according to claim 1, which is characterized by comprising the following steps: in step S106, the screened areas are combined to obtain the identified point cloud with significant error, which specifically includes: combining according to the adjacency relation of the screened areas, combining the screened areas with adjacency relation into a point cloud which is used as the point cloud of the same significant error point, and not combining the screened areas without adjacency relation to be used as the point cloud of different significant error points.
8. The method for identifying the point cloud normal and curvature change-based double-constraint surface error salient point according to claim 1, which is characterized by comprising the following steps: in S107, boundary extraction is performed on the identified point cloud of the error significant point, so as to obtain position information of the error significant point, which specifically includes: setting a boundary K neighbor search parameter K2 according to the error significant point cloud normal and the kd-tree dividing result to obtain a boundary point cloud of an error significant point; calculating the center of gravity P of the error salient point by using a boundary point cloud of the error salient point and the length and width a and b of the minimum bounding rectangle; the position information of the error significant point is the center of gravity P and the length and width a and b of the minimum bounding rectangle.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010675830.XA CN111986308B (en) | 2020-07-14 | 2020-07-14 | Point cloud normal and curvature change based double-constraint surface error salient point identification method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010675830.XA CN111986308B (en) | 2020-07-14 | 2020-07-14 | Point cloud normal and curvature change based double-constraint surface error salient point identification method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111986308A CN111986308A (en) | 2020-11-24 |
CN111986308B true CN111986308B (en) | 2023-09-01 |
Family
ID=73439483
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010675830.XA Active CN111986308B (en) | 2020-07-14 | 2020-07-14 | Point cloud normal and curvature change based double-constraint surface error salient point identification method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111986308B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112329179A (en) * | 2020-11-27 | 2021-02-05 | 华中科技大学 | Part feature rapid identification method based on point cloud data |
CN113486741B (en) * | 2021-06-23 | 2022-10-11 | 中冶南方工程技术有限公司 | Stock yard stock pile point cloud step identification method |
CN114627020B (en) * | 2022-03-18 | 2023-06-20 | 易思维(杭州)科技有限公司 | Method for removing reflection noise point of curved surface workpiece |
CN115690120B (en) * | 2022-06-06 | 2024-01-26 | 如你所视(北京)科技有限公司 | Method, apparatus, device, medium and program product for generating three-dimensional grid |
CN116071274A (en) * | 2023-03-23 | 2023-05-05 | 成都工业学院 | Cross-plane hole repairing method and system based on point cloud region growth |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106023298A (en) * | 2016-06-22 | 2016-10-12 | 山东理工大学 | Point cloud rigid registration method based on local Poisson curved surface reconstruction |
CN107274422A (en) * | 2017-05-08 | 2017-10-20 | 燕山大学 | A kind of point cloud edge extracting method combined based on normal information and K neighborhood search |
CN109409437A (en) * | 2018-11-06 | 2019-03-01 | 安徽农业大学 | A kind of point cloud segmentation method, apparatus, computer readable storage medium and terminal |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8493403B2 (en) * | 2009-09-04 | 2013-07-23 | Hewlett-Packard Development Company, L.P. | Color transformation sample point identification |
-
2020
- 2020-07-14 CN CN202010675830.XA patent/CN111986308B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106023298A (en) * | 2016-06-22 | 2016-10-12 | 山东理工大学 | Point cloud rigid registration method based on local Poisson curved surface reconstruction |
CN107274422A (en) * | 2017-05-08 | 2017-10-20 | 燕山大学 | A kind of point cloud edge extracting method combined based on normal information and K neighborhood search |
CN109409437A (en) * | 2018-11-06 | 2019-03-01 | 安徽农业大学 | A kind of point cloud segmentation method, apparatus, computer readable storage medium and terminal |
Also Published As
Publication number | Publication date |
---|---|
CN111986308A (en) | 2020-11-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111986308B (en) | Point cloud normal and curvature change based double-constraint surface error salient point identification method | |
CN110060338B (en) | Prefabricated part point cloud identification method based on BIM model | |
Rabbani et al. | Segmentation of point clouds using smoothness constraint | |
CN109658381B (en) | Method for detecting copper surface defects of flexible IC packaging substrate based on super-pixels | |
JP4926116B2 (en) | Image inspection device | |
CN106548520A (en) | A kind of method and system of cloud data denoising | |
CN107358636B (en) | Loose defect image generation method based on texture synthesis | |
CN111354047B (en) | Computer vision-based camera module positioning method and system | |
CN115235375A (en) | Multi-circle characteristic parameter measuring method, detecting method and device for cover plate type workpiece | |
Yogeswaran et al. | 3d surface analysis for automated detection of deformations on automotive body panels | |
CN111783722B (en) | Lane line extraction method of laser point cloud and electronic equipment | |
CN116523898A (en) | Tobacco phenotype character extraction method based on three-dimensional point cloud | |
CN117115012A (en) | Road surface point cloud marking segmentation denoising method | |
CN117854060B (en) | Tunnel rock body planar crack identification method and system based on deep learning | |
Zhang et al. | Automatic extraction of DTM from low resolution DSM by twosteps semi-global filtering | |
CN117635615A (en) | Defect detection method and system for realizing punching die based on deep learning | |
CN106815851B (en) | A kind of grid circle oil level indicator automatic reading method of view-based access control model measurement | |
CN116452826A (en) | Coal gangue contour estimation method based on machine vision under shielding condition | |
CN117292181A (en) | Sheet metal part hole group classification and full-size measurement method based on 3D point cloud processing | |
Omidalizarandi et al. | Segmentation and classification of point clouds from dense aerial image matching | |
CN115810153A (en) | Relative entropy and semantic constraint-based highway point cloud extraction method | |
JP2005165969A (en) | Image processor and method | |
CN115880228A (en) | Multi-defect merging method and device, computer equipment and storage medium | |
CN110334664B (en) | Statistical method and device for alloy precipitated phase fraction, electronic equipment and medium | |
CN117409024B (en) | High-robustness high-voltage power line segmentation method for sparse point cloud and application thereof |
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 |