TW201621813A - Method and system for processing point clouds of an object - Google Patents

Abstract

The present invention provides a method and a system for processing point clouds of an object. The system comprises one or more modules. The one or more modules retrieve data of the point clouds of the object from a storage device of an electronic device. An average value of distances between any two point clouds and a curvature parameter are obtained from a user. The one or more modules determine an area of the point clouds that needs to be processed, and determine a minimum encasing box of the determined area. A plurality of points is selected from a region between the determined area and the minimum encasing box, and a curved surface is constructed based on the plurality of points. The one or more modules interpolate on the curved surface based on the average value to obtain one or more discrete points, and insert the discrete points to the determined area of the point clouds. One or more unorganized points of the determined area are determined based on the average value and the curvature parameter, and further deleted from the determined area. When the point clouds do not comprise other determined areas to process, the one or more modules generate smooth treatment on the point clouds.

Description

Point cloud repair method and system

The present invention relates to a point cloud processing method and system, and more particularly to a method and system for repairing a missing point cloud.

When the 3D scanning device scans the product to be measured to generate a point cloud, due to the reason of the scanning device itself or other external factors, the position that is difficult to scan by the three-dimensional scanning device and the position where the marking point is attached may cause a missing point cloud. In the existing point cloud processing system, the point cloud patch generally fits the missing surface through a simple point cloud, and fills the missing point with the point on the surface. This kind of point cloud repair method is easy to cause obvious filling marks in the missing place, and may cause distortion and deformation of the point cloud.

In view of the above, it is necessary to provide a point cloud repair method and system, which can realize high-precision repair of missing point clouds, which ensures the naturalness of the filling and ensures that the shape of the product to be measured does not change.

A point cloud repairing method is applied to an electronic device, the method comprising the steps of: (a) importing point cloud data of a product, and receiving a point spacing and a curvature filtering parameter input by a user; (b) in the point cloud data Determining a to-be-repaired area and determining a minimum bounding box of the area to be repaired; (c) selecting a preset number of points outside the to-be-repaired area and fitting the selected point to a B-spline surface (d) interpolating on the B-spline surface according to the point spacing to obtain a plurality of discrete points, and filling the plurality of discrete points into the area to be repaired; (e) filtering according to the point spacing and curvature a parameter, determining and deleting a clutter point of the to-be-repaired area; (f) determining whether there are other areas to be repaired in the point cloud data, and returning other area to be repaired in the point cloud data, returning Step (b); and when there is no other area to be repaired in the point cloud data, smoothing the filled point cloud data.

A point cloud repairing system, running in an electronic device, the system comprising: a data receiving module, configured to import point cloud data of a product, and receive a point spacing and a curvature filtering parameter input by a user; and an area determining module, configured to: When there is an area to be repaired in the point cloud data, a to-be-repaired area is determined in the point cloud data, and a minimum bounding box of the area to be repaired is determined; a surface fitting module is used in the area to be repaired And selecting a preset number of points in the minimum bounding box to fit the selected point into a B-spline surface; and a hole filling module for interpolating on the B-spline surface according to the point spacing to obtain multiple discretes Pointing, and filling the plurality of discrete points into the to-be-repaired area; the filtering module is configured to determine and delete the cluttered points of the to-be-repaired area according to the point spacing and curvature filtering parameters; The method is configured to determine whether there is another area to be repaired in the point cloud data, and a smoothing processing module, configured to: when there is no other area to be repaired in the point cloud data, the filled point is Data smoothed.

Compared with the prior art, the point cloud repairing method and system provided by the present invention can realize high-precision repair of the missing point cloud, which ensures the naturalness of the filling and ensures that the shape of the product to be measured does not change.

1 is a hardware architecture diagram of a preferred embodiment of a point cloud patching system of the present invention.

2 is a functional block diagram of a preferred embodiment of the point cloud patching system of the present invention.

3 is a flow chart showing the operation of a preferred embodiment of the point cloud repairing method of the present invention.

4 is an exemplary diagram of a cubic B-spline surface generated by the present invention.

Referring to Figure 1, there is shown a hardware architecture diagram of a preferred embodiment of the point cloud patching system of the present invention. The point cloud patching system 10 runs in the electronic device 100, and the electronic device 100 can be a computer, a server, a Programmable Logic Controller (PLC), a measuring machine, or other computing device capable of computing power. .

The electronic device 100 further includes components such as a processor 20, a storage unit 30, and a display unit 40. Components 10-40 of electronic device 100 communicate via a system bus.

The processor 20 is configured to execute the point cloud patching system 10 and various types of software installed in the electronic device 100, such as an operating system. The storage unit 30 can be a hard disk, or other type of memory card or storage device. The storage unit 30 is configured to store various types of materials, for example, point cloud data of different products and for storing data set, received, and processed by the point cloud patching system 10. The display unit 40 is used to display various types of visualized materials.

Referring to FIG. 2, it is a functional module diagram of a preferred embodiment of the point cloud patching system of the present invention. In the embodiment, the point cloud repairing system 10 includes a plurality of functional modules, namely: a data receiving module 11, a region determining module 12, a surface fitting module 13, a hole filling module 14, and a filtering module. The group 15, the determination module 16, and the smoothing module 17 are provided. The module referred to in the present invention refers to a series of computer programs that can be executed by the processor 20 and that can perform fixed functions, which are stored in the storage unit 30. In the present embodiment, the functions of the respective modules will be specifically described in the flowchart of FIG.

Referring to FIG. 3, it is a flow chart of a preferred embodiment of the point cloud repairing method of the present invention.

In step S310, the data receiving module 11 imports the point cloud data of the product and receives the parameters input by the user. The point cloud data includes, but is not limited to, a three-dimensional coordinate of a point, an identifier of a point, and a total number of point clouds. The point cloud data may be imported from the storage unit 30 of the electronic device 100, or may be imported from other external storage devices or point cloud scanning devices.

In this embodiment, the input parameters include, but are not limited to, a dot pitch and a curvature filter parameter. The dot pitch may be determined according to the digit input by the user, or the average value of the distance between all adjacent points in the point cloud data may be used as the dot pitch (also referred to as a point average pitch). In addition, in other embodiments, the dot pitch may also be dynamically generated according to a preset rule.

In this embodiment, the curvature filtering parameter is used to filter a point where the curvature changes too much. The curvature filtering parameter is set by the system or user, such as 0.8. It should be noted that the curvature of each point in the point cloud data is a value of 0 to 1.

In step S320, the area determining module 12 determines a to-be-repaired area in the point cloud data, and determines a minimum bounding box of the area to be repaired. The minimum bounding box may be, but is not limited to, a minimum outsourcing rectangle of the area to be repaired.

In this embodiment, the area to be repaired is determined by receiving a user's selection, or is determined by the area determining module 12 according to a preset method. For example, the preset method is: the area determining module 12 determines whether the area of the missing point cloud data in the point cloud data is larger than a preset area, and the area of the missing point cloud data in the point cloud data is larger than the area. When the preset area is described, the hole is determined to be the area to be repaired. The predetermined area is determined according to an area of an object (for example, a marker point) that causes missing point cloud data, such as 80% of an area of an object that causes missing point cloud data.

In step S330, the surface fitting module 13 selects a preset number of points outside the to-be-repaired area and the minimum bounding box, and fits the selected points into a B-spline surface by using a preset mathematical method. The predetermined mathematical method may be a Newton iteration method, or other mathematical methods employed in the prior art. The Newton iteration method can speed up the fitting of B-spline surfaces.

The B-spline surface can be a cubic B-spline surface to achieve maximum retention of product features, and the cubic B-spline surface is fitted as shown in FIG. 4 .

In this embodiment, the preset number of points are evenly and discretely distributed outside the to-be-repaired area and located in the minimum bounding box, for example, selecting a point every other first preset distance to obtain a A preset number of points, the first preset distance being half of the total side length of the rectangle. The preset number of points is preset by the user or the system, such as 5.

Step S340, the hole filling module 14 interpolates on the B-spline surface according to the dot pitch to obtain a plurality of discrete points, and fills the plurality of discrete points into the to-be-repaired area.

In the present embodiment, the hole filling module 14 takes an isoparametric curve at equal intervals in the x1 direction of the B-spline surface according to the dot pitch, and then equally spaced each of the isoparametric curves in the x2 direction according to the dot pitch. Point and use the interval points as the discrete points.

In other embodiments, the hole filling module 14 also utilizes the curvature of each equally spaced point to obtain discrete points. For example, after obtaining the interval points, the hole filling module 14 determines whether the curvature of each equal interval point is greater than a preset curvature value, and when the curvature of the interval points is greater than the preset curvature value, the interval is The point is moved to the left by a second preset distance according to the x1 direction. When the curvature of the interval points is not greater than the preset curvature value, the positions of the equal interval points remain unchanged, and the equal interval points constitute the discrete point. The second preset distance=(curvature-preset curvature value)*point spacing, the preset curvature value is preset by the user or the system, such as 0.5. This method of obtaining discrete points by using curvature change as a weight is more in line with the distribution characteristics of the point cloud (where the curvature changes greatly, the point cloud density is large), so that the patching of the point cloud is more natural.

The curvature of each of the equally spaced points is used to characterize the degree of deformation of each equally spaced point on the B-spline surface. The curvature of each of the equally spaced points can be derived using the curvature found on the B-spline surface as is known in the art.

In step S350, the filtering module 15 determines and deletes the messy points of the to-be-repaired area according to the point spacing and the curvature filtering parameter. The clutter points include, but are not limited to, points having a curvature greater than the curvature filtering parameter and isolated points having no adjacent points.

In this embodiment, the filter module 15 determines whether there are other points in the preset interval within the to-be-repaired area. When a certain point does not have other points within the preset interval, The point is determined to be an isolated point with no neighboring points. The preset spacing is preset by the system or the user, such as a point average spacing of 2 times.

In step S360, the determining module 16 determines whether there are other areas to be repaired that need to be repaired in the point cloud data. When there are other areas to be repaired in the point cloud data that need to be repaired, the process returns to step S320. When there is no other area to be repaired in the point cloud data that needs to be repaired, step S370 is performed.

In step S370, the smoothing processing module 17 performs smoothing processing on the filled point cloud data. When the smoothing process is performed, the smoothing processing module 17 determines the normal vector of each point, and iterates by moving the points along the direction of the normal vector by the moving minimum distance method until the points satisfy the accuracy of the smoothing requirement and then stops. The accuracy that satisfies the smoothing requirement is preset by the user or the system, such as one tenth of the average dot pitch.

In summary, the present invention complies with the requirements of the invention patent and submits a patent application according to law. However, the above description is only the preferred embodiment of the present invention, and the scope of the present invention is not limited to the above embodiments, and equivalent modifications or variations made by those skilled in the art in accordance with the spirit of the present invention are It should be covered by the following patent application.

100‧‧‧Electronic devices

10‧‧‧ Point Cloud Patching System

11‧‧‧ Data receiving module

12‧‧‧Regional Determination Module

13‧‧‧Surface fitting module

14‧‧‧ hole filling module

15‧‧‧Filter module

16‧‧‧Judgement module

17‧‧‧Smoothing module

20‧‧‧ processor

30‧‧‧ storage unit

40‧‧‧Display unit

no

10‧‧‧ Point Cloud Patching System

11‧‧‧ Data receiving module

12‧‧‧Regional Determination Module

13‧‧‧Surface fitting module

14‧‧‧ hole filling module

15‧‧‧Filter module

16‧‧‧Judgement module

17‧‧‧Smoothing module

Claims (10)

A point cloud repairing method is applied to an electronic device, the method comprising:
Data receiving step: importing the point cloud data of the product, and receiving the point spacing and curvature filtering parameters input by the user;
a region determining step: determining a to-be-repaired area in the point cloud data, and determining a minimum bounding box of the to-be-repaired area;
a surface fitting step: selecting a preset number of points outside the area to be repaired and in the minimum bounding box, and fitting the selected point into a B-spline surface;
a hole filling step: interpolating on the B-spline surface according to the dot pitch to obtain a plurality of discrete points, and filling the plurality of discrete points into the to-be-repaired region;
a filtering step: determining and deleting a clutter point of the to-be-repaired area according to the point spacing and curvature filtering parameters;
a determining step: determining whether there are other areas to be repaired in the point cloud data, and returning an area determining step when there are other areas to be repaired in the point cloud data; and a smoothing processing step: when the point cloud When there is no other area to be repaired in the data, the filled point cloud data is smoothed. The point cloud repairing method described in claim 1, wherein the area to be repaired is determined by the following method:
Determining whether the area of the missing point cloud data in the point cloud data is larger than a predetermined area, and determining that the hole is the area to be repaired when the area of the missing point cloud data in the point cloud data is larger than the predetermined area . For the point cloud repair method described in claim 1, the method of fitting the selected point to the B-spline surface is a Newton iteration method. The point cloud repairing method of claim 1, wherein the plurality of discrete points are obtained by:
Obtaining an isoparametric curve at equal intervals in the x1 direction of the B-spline surface according to the dot pitch, and taking equal intervals according to the dot pitch in the x2 direction for each isoparametric curve;
Determining whether the curvature of each equally spaced point is greater than a preset curvature value;
When the curvature of the interval points is greater than the preset curvature value, the interval points are moved to the left by a preset distance according to the x1 direction; and the equal interval points are used as the discrete points. The point cloud repairing method of claim 1, wherein the clutter point comprises a point having a curvature greater than the curvature filtering parameter and an isolated point having no adjacent point. A point cloud patching system running in an electronic device, the system comprising:
a data receiving module, configured to import point cloud data of the product, and receive a point spacing and a curvature filtering parameter input by the user;
a region determining module, configured to determine a to-be-repaired area in the point cloud data when the to-be-repaired area exists in the point cloud data, and determine a minimum bounding box of the to-be-repaired area;
a surface fitting module, configured to select a preset number of points outside the area to be repaired and in the minimum bounding box, and fit the selected point into a B-spline surface;
a hole filling module, configured to interpolate on the B-spline surface according to the dot pitch to obtain a plurality of discrete points, and fill the plurality of discrete points into the to-be-repaired area;
a filtering module, configured to determine, according to the point spacing and the curvature filtering parameter, the disordered point of the to-be-repaired area;
a judging module, configured to determine whether there is another area to be repaired in the point cloud data; and a smoothing processing module, configured to: when there is no other area to be repaired in the point cloud data, the filled module Point cloud data is smoothed. The point cloud repairing system of claim 6, wherein the area determining module determines the area to be repaired by:
Determining whether the area of the missing point cloud data in the point cloud data is larger than a predetermined area, and determining that the hole is the area to be repaired when the area of the missing point cloud data in the point cloud data is larger than the predetermined area . For the point cloud repair system described in claim 6, the method of fitting the selected point to the B-spline surface is a Newton iteration method. The point cloud patching system of claim 6, wherein the hole filling module obtains a plurality of discrete points by:
Obtaining an isoparametric curve at equal intervals in the x1 direction of the B-spline surface according to the dot pitch, and taking equal intervals according to the dot pitch in the x2 direction for each isoparametric curve;
Determining whether the curvature of each equally spaced point is greater than a preset curvature value;
When the curvature of the interval points is greater than the preset curvature value, the interval points are moved to the left by a preset distance according to the x1 direction; and the equal interval points are used as the discrete points. The point cloud patching system of claim 6, wherein the clutter point comprises a point having a curvature greater than the curvature filtering parameter and an isolated point having no adjacent point.