JP5948480B1 - System, method and computer software program for displaying / manipulating 2D or 3D shape information - Google Patents

Abstract

Kind Code: A1 A shape data processing system capable of acquiring a thinning point group with little bias from a high density point group is provided. An input means for inputting and processing information of a point cloud before thinning obtained by measuring a real space, a point cloud text file processing section, and a point for performing the thinning processing of the inputted point cloud. And a group data thinning processing unit 10. The point cloud data thinning processing unit 10 sequentially divides a predetermined target area including the point cloud into two, and generates a tree structure data divided into a number of areas (nodes) as it goes down. The node 10a and a node for performing thinning processing in the tree structure data are determined for each branch of the tree structure, and the determined node group is selected as a divided region group to be acquired as a point group after the thinning, and is selected. In addition, each node has a thinning unit 10b that selects one or more points representing the divided area represented by the node as a representative point, and acquires a total of the selected representative point group as a thinned point group. [Selection] Figure 1

Description

  The present invention relates to a system for displaying and operating two-dimensional or three-dimensional shape information, particularly point cloud data.

  Conventionally, there is a system that measures a predetermined space using, for example, a three-dimensional space scanner and outputs the measurement result as point cloud data.

  The point cloud data is assumed to be used for applications such as, for example, creation of a three-dimensional model of a part, quality control by measurement, visualization processing, animation, rendering, and mass customization.

JP 2006-208262 A

  By the way, when utilizing point cloud data, it is assumed that the density of a point cloud required according to a use differs. In other words, while there is an application that needs to prepare point cloud data that satisfies a high density, for example, when the display is aimed only for display without requiring model operation, the display performance deteriorates if the number of point clouds is too large. May be a problem.

  For this reason, it is assumed that the number of point clouds is reduced by thinning the point clouds from the high-density point clouds by a predetermined method. However, the distribution density of the point clouds is not uniform over the entire space, and the adjacent point clouds are simply In processing such as thinning out one of the points, there is a possibility that the point cloud after thinning is biased and the number of point clouds cannot be reduced while maintaining the shape characteristics.

  The present invention has been made in view of such circumstances, and it is an object of the present invention to provide a shape data processing system capable of acquiring a thinning point group with a small deviation from a high density point group.

  In order to solve the above-mentioned problem, according to a first main aspect of the present invention, a system for displaying / manipulating two-dimensional or three-dimensional shape information before thinning obtained by measuring a real space Means for inputting point cloud information, means for generating tree structure data in which a predetermined target area including the point cloud is sequentially divided into two parts and divided into a number of areas (nodes) as it goes down, and a tree structure Means for determining a node for performing thinning processing in the data for each branch of the tree structure; means for selecting the determined node group as a divided region group to be acquired as a point group after thinning; and each of the selected nodes And a means for selecting one or more points representing a divided area represented by the node as a representative point in a node, and obtaining a total of the selected representative point group as a thinned-out point group. Is .

  According to such a configuration, an area including a point cloud can be divided into each node belonging to the tree structure data, a representative point can be selected from a predetermined node, and the total of the representative points can be used as a thinned point cloud. it can. This makes it possible to acquire a thinning point group with less bias from the high density point group.

  That is, in the region where the point cloud density is relatively high in the inputted point cloud data, the point cloud density is relatively high even in the point cloud data after thinning, and the original point cloud data is relatively In an area where the density of point clouds is low, the density of point clouds is relatively low even in the point cloud data after thinning.

  Further, according to one embodiment of the present invention, the system further includes means for inputting the number of point groups after thinning, and the means for determining a node to perform the thinning processing is the point group before the thinning. Based on the ratio of the inputted point cloud number after thinning to the number, the node when the number obtained by multiplying the number of points included in each node of the tree structure by the ratio is smaller than a predetermined number is set as a node for performing the thinning process To decide. In this case, the predetermined number is preferably 1. Further, the means for designating the number of point clouds after the thinning designates a small number of thinning point groups for display or animation display accompanied by dynamic viewpoint change according to the display purpose of the point cloud data. It is desirable to specify a larger number of thinning point groups for display and measurement.

  According to another embodiment of the present invention, the system further includes point cloud data output means for outputting the acquired thinned point cloud data in an intermediate format for storing drawing information.

  According to a second main aspect of the present invention, there is provided a method for displaying / manipulating two-dimensional or three-dimensional shape information executed by a computer, the thinning obtained by measuring a real space on the computer. The step of inputting the information of the previous point cloud, and the computer sequentially generates a tree structure data that is divided into two regions (nodes) as the predetermined target region including the point cloud is sequentially divided into two parts A step in which a computer determines a node for performing thinning processing in the tree structure data for each branch of the tree structure, and a computer in which the determined node group is an acquisition target of a point group after thinning The process of selecting as an area group, and the computer selects and selects one or more points representing the divided area represented by the node in each of the selected nodes as a representative point. Method characterized by a step of obtaining the totality of the representative point group as a decimation rear point group is provided.

  According to a third main aspect of the present invention, there is provided a computer software program stored in a storage medium and causing a computer to display and operate two-dimensional or three-dimensional object data. The computer measures a real space. A tree obtained by dividing the predetermined target area including the point group into two parts in succession and dividing it into a number of areas (nodes) as it goes down. Means for generating structure data, means for causing the computer to determine a node for thinning processing in the tree structure data for each branch of the tree structure, and for the computer to determine the determined node group for acquisition of the point cloud after thinning Means for selecting as a divided area group, and a computer having one or more of the divided areas represented by the node in each of the selected nodes. Was selected as the representative point in terms of a computer software program, characterized in that it comprises a means for obtaining the totality of sorted representative point group as a decimation rear point group is provided.

  Other technical features of the present invention which are not described herein will be clarified in the description of the following embodiments and attached drawings.

FIG. 1 is a schematic configuration diagram showing an embodiment of the present invention. FIG. 2 is a flowchart showing processing similarly. FIG. 3 is a conceptual diagram showing point cloud data before thinning and point cloud data after thinning. FIG. 4 is a schematic diagram showing a tree structure of BVH. FIG. 5 is a conceptual diagram showing nodes that have undergone thinning processing according to the point group BVH and rough point groups after thinning (representative points).

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a schematic configuration diagram showing a point cloud data thinning system 1 according to the present invention.

  In this system 1, a program storage unit 6 and a data storage unit 7 are connected to a bus 5 to which a CPU 2, a RAM 3, an input / output interface 4 and the like are connected. The program storage unit 6 includes a point cloud text file processing unit 8, a rough point cloud number input unit 9, a point cloud data thinning processing unit 10, a point cloud data file output unit 11, and an IGES file output unit 12. Stored. The data storage unit 7 stores a point cloud text file 13, a point cloud data file 14, and an IGES file 15. Further, the point cloud data file 14 is composed of rough point cloud data 14a and detailed point cloud data 14b. Here, “rough point cloud” represents a point cloud remaining after the thinning process, and “detailed point cloud” represents a point cloud thinned out by the thinning process.

  Here, the program storage unit 6 and the data storage unit 7 are actually storage media such as a hard disk. Each of the components 8 to 12 is actually a computer software program, and functions as each component of the present invention by being called, expanded and executed on the RAM 3 by the CPU 2. .

  Hereinafter, the configuration and operation of each of the components 8 to 12 will be described together with the actual operation with reference to the flowchart of FIG.

  First, in step S1 (FIG. 2), the point cloud text file processing unit 8 reads the point cloud text file 13 and converts it into temporary point cloud data that can be processed by the point cloud data thinning processing unit 10. This temporary point cloud data is expanded in the RAM 3. Here, the point cloud text file 13 represents, in a text format, coordinate values and color information of each point on the space measured by, for example, a three-dimensional space scanner 16 connected to the input / output interface 4. Point cloud information. FIG. 3A shows the temporary point cloud data 19 converted by the point cloud text file processing unit 8 and drawn on the display as the output means 17 connected to the input / output interface 4. It is. This system 1 is intended to generate rough point group data obtained by thinning out a desired number of points as shown in FIG. 3B from the point group data shown in FIG.

  In this embodiment, the point cloud text file 13 is assumed to be data output from the three-dimensional space scanner 16, but may be created by other methods.

  Next, in steps S2 to S3, the rough point group number input unit 9 designates the range of point cloud data to be thinned out (step S2) and the number of rough point groups (step S3). Here, the range of the point cloud data is specified by a user drawing a box on the display (output unit 17) using the mouse or the like as the input unit 18, and a point group included in the range (or a point group not included). This may be done by designating or by actually inputting a coordinate range as data. Also, all of the point cloud data included in the point cloud text file 13 may be targeted. Further, at this time, the point cloud text file processing unit 8 is obtained by performing processing such as rotating, moving, or enlarging the point cloud data on the output unit 17 based on an instruction from the input unit 18. It may be configured as follows.

  As the number of rough point groups specified in step S3, the number of point groups after thinning desired by the user may be specified, or may be specified as a percentage. When specifying with a specific number of point groups, the rough point group number input unit 9 acquires and displays the total number of point groups included in the selection range for user reference. Preferably it is.

  Further, the rough point cloud number input unit 9 determines the number of rough point groups (or a thinning ratio) when the purpose of use of the point cloud data is input from the input unit 18. It further has a portion 9a. This purpose-of-use discriminating unit 9a specifies, for example, a small number of thinned point groups for display or animation display with dynamic viewpoint change according to the display purpose of point cloud data, and for static display or measurement. Specify a larger number of thinning points.

  Next, in steps S4 to S6, the point cloud data thinning processor 10 performs a point cloud thinning process. As shown in FIG. 1, the point cloud data thinning processing unit 10 includes a BVH region division processing unit 10a that divides a region including a point cloud into a plurality of regions using a BVH technique, and points included in each of the divided regions. According to the number and density of the groups, the thinning unit 10b for thinning out the point groups at a constant rate (ratio of the number of rough point groups to the total number of point groups) is provided for each region.

  First, the BVH area division processing unit 10a divides a designated area (a group of points included therein) into a plurality of areas using the BVH technique (step S4).

  Here, BVH is an abbreviation for Bounding Volume Hierarchy, and is generally known as a technique for performing distance measurement in space at high speed, but is a new application in the present invention. It is used as a technique for obtaining a thinning point group with little bias from a high density point group. The BVH used in the present technology is hereinafter referred to as “point cloud BVH”.

  The principle of this technique is to construct tree structure data by sequentially dividing a space including elements in a three-dimensional space into two. Since this tree structure is an acyclic directed graph and is created by dividing the space into two, each node has either two children or no children at all. Usually, a node having no children is associated with the minimum unit of the target element.

  FIG. 4 is an example of BVH with a triangle in space as a target element. For simplicity, it is represented by a two-dimensional plane. The tree structure on the left is a schematic diagram of BVH. Here, the parent node of the BVH is called a node with a depth of 1, and as the generation goes down from there, the node group with a depth of 2 and a node group with a depth of 3 are called up one by one.

  Each node has a bounding box associated with it. Here, the “boundary box” is indicated by a dotted line on the right side of FIG. 4 and indicates the smallest rectangular parallelepiped including the target triangle. The right side of FIG. 4 shows an image of a triangle included in each bounding box and a rectangle representing the bounding box (since it is a two-dimensional plane).

  In this way, the region including the input point group is divided into a plurality of regions by the BVH technique. However, since the density of the point group is not constant over the entire region, the number of points included in each node differs even at the same depth.

  Next, the thinning unit 10b determines a node to perform the thinning process in order to perform a specific thinning process based on the total number of point groups and the number of rough point groups.

  That is, the following information can be acquired by using the point group BVH.

-A plurality of regions including point clouds in a three-dimensional space-Number and density of point clouds in each region The thinning unit 10b uses these pieces of information to determine which node performs thinning processing Each tree branch is determined by the following method.

  The ratio X of the number of rough point groups to the total number of point groups is obtained by designation in the rough point group number input unit 9.

  Tracing children from a node with a small depth to a node with a large depth. This is done by examining the branches one by one. At this time, the number of point groups included in the area corresponding to each node is confirmed.

  When the value obtained by multiplying the number of point groups by the ratio X of the number of rough point groups is equal to or less than 1, the corresponding node is set as a “node that performs a thinning process”.

  -The search is not made below the child nodes of the above-mentioned “node performing the thinning process”.

  Next, at each node that performs the thinning process, a point cloud having a ratio obtained by the ratio of the number of detailed point clouds to the total number of point clouds is thinned (step S6). One point group remains in each area, and the remaining point is the point closest to the center of the bounding box representing each area.

  FIG. 5 shows an example of a rough point group after the thinning process and a node that has performed the thinning process. For simplicity, it is represented by a two-dimensional plane. In this example, the nodes on the left side and the lower right side of FIG. 5 are determined as “nodes that perform thinning processing” by nodes having a smaller depth, and the nodes on the upper right side are “nodes that perform thinning processing” by nodes having a greater depth. As determined.

  By the above processing, thinning that satisfies the following conditions becomes possible.

  -The relative density of point clouds should not change before and after thinning. In other words, an area where the density of point clouds is high has a relatively high density for the entire point cloud even after thinning is performed, and an area where the density of point clouds is low is relatively low even after the thinning is executed. There is.

  ・ The number of rough points is almost the same as the value input from the outside.

  As a result, the point cloud data file 14 is output via the point cloud data file output unit 11 (step S7). The rough point group and the detailed point group obtained in step S6 are output as rough point group data 14a and detailed point group data 14b, respectively.

  The output rough display point cloud data 14a and detailed point cloud data 14b have the following properties.

  Arbitrary points included in the rough point group data 14a and the detailed point group data 14b have a one-to-one correspondence with the points included in the input one-time group data 19 in the three-dimensional space. Further, the coordinate values of the corresponding points in the three-dimensional space are the same.

  The number of point groups in the rough point group data 14 a is almost the value specified by the rough point group number input unit 9.

  The rough point cloud data 14a and the detailed point cloud data 14b are not necessarily point cloud data having a uniform density. In the input point-in-time group data 19 in the three-dimensional space, in the region where the point cloud density is relatively high, the point cloud density is relatively high also in the rough point cloud data 14a and the detailed point cloud data 14b. . On the contrary, in the point-time group data 19 in the three-dimensional space, in the region where the point cloud density is relatively low, the point cloud density is relatively low also in the rough point cloud data 14a and the detailed point cloud data 14b. .

In addition, it is possible to output a rough point group to the IGES file 15 via the IGES file output unit 12 (step S8). The IGES file 15 is an IGES file that is an intermediate format for storing drawing information.
This makes it easy to support the evaluation and editing of the point cloud data processed by this system using drawing editing software such as a CAD system.

  The example described above is merely one embodiment of the present invention, and various modifications can be made without departing from the scope of the present invention.

1 ... Point cloud data thinning system 2 ... CPU
3 ... RAM
DESCRIPTION OF SYMBOLS 4 ... Input / output interface 5 ... Bus 6 ... Program storage part 7 ... Data storage part 8-12 ... Configuration part 8 ... Point cloud text file processing part 9 ... Rough point cloud number input part 9a ... Point cloud use purpose discrimination | determination part 10 ... Point cloud data thinning processing unit 10a ... BVH region division processing unit 10b ... thinning unit 11 ... point cloud data file output unit 12 ... IGES file output unit 13 ... point cloud text file 14 ... point cloud data file 14a ... rough point cloud data 14b ... detailed point cloud data 15 ... IGES file 16 ... 3D space scanner 17 ... output means 18 ... input means 19 ... temporary point cloud data

Claims (15)

In a system that displays and manipulates 2D or 3D shape information,
Means for inputting the point cloud information before thinning obtained by measuring the real space;
Means for generating tree structure data that is divided into a plurality of regions (nodes) as the predetermined target region including the point group is sequentially divided into two and going downward;
Means for determining, for each branch of the tree structure, a node for performing thinning processing in the tree structure data;
Means for selecting the determined node group as a divided region group to be acquired for the point group after thinning;
Means for selecting one or more points representing the divided area represented by the node in each of the selected nodes as a representative point, and obtaining a total of the selected representative point group as a point group after thinning;
An arbitrary point included in the point group after thinning is a system in which a point corresponding to one-to-one exists in the point group before thinning. The system of claim 1, wherein
Furthermore, it has means for inputting the number of point clouds after thinning,
The means for determining the node to perform the thinning process is based on the ratio of the inputted point group number after thinning to the number of point groups before the thinning process, and the number of points included in each node of the tree structure is multiplied by the above ratio. A system characterized in that a node when a number becomes smaller than a predetermined number is determined as a node for performing a thinning process. The system of claim 2, wherein
The system is characterized in that the predetermined number is one. The system of claim 2, comprising:
The means for designating the number of point groups after the thinning designates a small number of thinned point groups for display or animation display with dynamic viewpoint change according to the display purpose of the point cloud data, A system that specifies a larger number of thinning points for measurement. The system of claim 1, wherein
The system further comprises point cloud data output means for outputting the acquired thinned point cloud data in an intermediate format for storing drawing information. A method of displaying and operating two-dimensional or three-dimensional shape information executed by a computer,
Inputting the information of the point cloud before thinning obtained by measuring the real space to the computer;
A step of generating a tree structure data in which a computer sequentially divides a predetermined target region including the point cloud into two and divides into a plurality of regions (nodes) as going down
A step in which a computer determines a node for performing a thinning process in the tree structure data for each branch of the tree structure;
A step of selecting the determined node group as a divided region group to be acquired as a point group after thinning;
A step of selecting, as a representative point, one or more points representing the divided area represented by the node in each of the selected nodes, and obtaining a total of the selected representative point group as a thinned point group; And
An arbitrary point included in the point group after thinning is a point corresponding to one-to-one in the point group before thinning. The method of claim 6 wherein:
Furthermore, the computer has a step of inputting the number of point clouds after thinning,
The step of determining a node to perform the thinning process is performed by the computer based on the ratio of the inputted point group number after thinning to the number of point groups before the thinning process. A method in which a node when the number multiplied by becomes smaller than a predetermined number is determined as a node to perform a thinning process. The method of claim 7, wherein
The predetermined number is one. The method of claim 7, comprising:
The step of designating the number of point clouds after thinning designates a small number of thinning point groups for display or animation display accompanied by dynamic viewpoint change according to the display purpose of the point cloud data. A method characterized by specifying a larger number of thinning points for accurate display and measurement. The method of claim 6, wherein
The computer further comprises a point cloud data output step of outputting the acquired point cloud data after thinning out in an intermediate format for storing drawing information. A computer software program stored in a storage medium and causing a computer to display and operate two-dimensional or three-dimensional object data,
Computer
Means for inputting the point cloud information before thinning obtained by measuring the real space;
Means for generating tree structure data that is divided into a plurality of regions (nodes) as the predetermined target region including the point group is divided into two in order and going downward;
Means for determining a node for thinning processing in tree structure data for each branch of the tree structure;
Means for selecting the determined node group as a divided region group to be acquired as a point group after thinning;
Means for selecting one or more points representing the divided area as the node in each of the selected nodes as a representative point, and obtaining the total of the selected representative point group as a point group after thinning
As a function,
An arbitrary point included in the point group after thinning is a computer software program characterized in that a point corresponding to one-to-one exists in the point group before thinning. The computer software program according to claim 11, wherein
Furthermore, the computer is intended to function as means for inputting a point number group after thinning,
The means for determining the node to perform the thinning process is based on the ratio of the inputted point group number after thinning to the number of point groups before the thinning process, and the ratio of the computer to the number of points included in each node of the tree structure computer software program, wherein the number multiplied by is intended to function in so that is determined as a node for performing thinning processing nodes when smaller than the predetermined number. The computer software program according to claim 12, wherein
The predetermined number is 1. A computer software program, characterized in that A computer software program according to claim 12, comprising:
Means for designating a point number group after the decimation, the computer, in accordance with the display object of the point group data, to specify the number of small decimation point group on the display and animation display with dynamic viewpoint change, static computer software program, characterized in that the Do display and measurement is also of a to function so as to specify a greater number of decimation point group. The computer software program according to claim 11,
A computer software program for causing a computer to function as point cloud data output means for outputting the acquired point cloud data after thinning out in an intermediate format for storing drawing information.

Images