Merged point cloud data generation method, point cloud as-built drawing generation method, program, and computer readable recording medium

Abstract

To provide a highly accurate merged point cloud data generation method based on point cloud data obtained by surveying traditional architecture, such as shrine and temple buildings, using a 3D scanner, a program, and a recording medium.SOLUTION: A method includes the steps of: acquiring first point cloud data obtained by surveying a first space, which is an open space of traditional architecture, using a 3D scanner, and second point cloud data obtained by surveying a second space which is obscure and defined as a space different from the first space in the traditional architecture, using the 3D scanner; creating a joint group by associating small groups with second point cloud data located at positions corresponding to the small groups, the small groups being formed by partitioning a large space set for each unit of acquiring the first point cloud data into a predetermined number of units; determining a position of the joint group in the large space; and generating merged point cloud data by merging the first point cloud data with the second point cloud data based on the determined position.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to a method, program, and computer-readable recording medium for generating synthetic point cloud data for point cloud data obtained by surveying traditional architecture such as shrines and temples using a 3D scanner. [Background technology]

[0002] Surveying using a 3D scanner is being considered in places where measurements using a convex (tape measure) are time-consuming or difficult. For example, surveying using a 3D scanner is being carried out to create a layout plan when adding arch trusses to large-space architectural structures such as dome-shaped stadiums (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-117041 Summary of the Invention [Problem to be solved by the invention]

[0004] However, in Patent Document 1, the building area is 50,000 m 2 The study focused on large-space architectural structures exceeding 100m2, but did not take into account traditional architecture such as temples and shrines, or old houses. 2 From about several hundred meters 2 At most, it is about 1000m 2 Furthermore, the target building in Patent Document 1 was constructed recently, blueprints exist, and the survey using a 3D scanner was carried out in an open space that is visible to users.

[0005] On the other hand, many traditionally constructed temples, shrines, and traditional houses have high cultural value, and therefore require blueprints for restoration. However, sometimes construction blueprints do not exist, and even if they do exist, they may not match the current state due to additions or renovations. Furthermore, when creating blueprints, it is necessary to survey spaces that are not visible to users, such as attics, crawl spaces, and ceilings, which are separate from the open spaces used by users (priests, worshippers, residents, etc.). However, because attics, crawl spaces, and ceilings in traditional architecture are not intended for use by users, they are often confined spaces, making surveying them difficult. Even if surveying were possible, it was difficult to generate accurate blueprints from the obtained data using existing algorithms. [Means for solving the problem]

[0006] The object of the present invention is to provide a method, program, and computer-readable recording medium for generating synthetic point cloud data that can generate highly accurate synthetic point cloud data based on point cloud data obtained by surveying traditional architecture such as shrine and temple buildings using a 3D scanner.

[0007] As a result of extensive research, the inventors have discovered that the above problems can be solved by performing a predetermined process on point cloud data, and have completed the present invention.

[0008] That is, according to the present invention, (1) A method for generating composite point cloud data, the method comprising the steps of: acquiring first point cloud data obtained by surveying a first space, which is an open space in a traditional building, with a 3D scanner; and acquiring second point cloud data obtained by surveying a second space, which is partitioned as a space separate from the first space in the traditional building and is not visible to the public, with a 3D scanner; creating joint groups by associating small groups, which are obtained by dividing a large space set for each acquisition unit when acquiring the first point cloud data, with the second point cloud data at positions corresponding to the small groups; determining the position of the joint group in the large space; and generating composite point cloud data by synthesizing the first point cloud data and the second point cloud data based on the determined positions; (2) The method for generating composite point cloud data according to (1), wherein the second point cloud data is data obtained by a step of confirming the load capacity of an introduction path of a 3D scanner into the second space, a step of introducing the 3D scanner into the second space when the load capacity of the introduction path does not hinder the introduction of the 3D scanner, and a step of measuring three-dimensional measurement data of the second space using the 3D scanner; (3) A method for generating a point cloud current state preservation drawing, comprising the step of extracting a line drawing based on the composite point cloud data according to (1) or (2); (4) A program for causing a computer to function as a means for acquiring first point cloud data obtained by surveying a first space, which is an open space in a traditional building, with a 3D scanner, and second point cloud data obtained by surveying a second space, which is partitioned as a space separate from the first space in the traditional building and is not visible to the public, with a 3D scanner; a means for creating a joint group by combining small groups obtained by dividing a large space into a predetermined number of units set for each acquisition unit when acquiring the first point cloud data, and the second point cloud data at positions corresponding to the small groups; a means for determining the position of the joint group in the large space; and a means for generating synthesized point cloud data by synthesizing the first point cloud data and the second point cloud data based on the determined positions. (5) The program according to (4), wherein the second point cloud data is data obtained by a step of confirming the load capacity of an introduction path of a 3D scanner into the second space, a step of introducing the 3D scanner into the second space when the load capacity of the introduction path does not hinder the introduction of the 3D scanner, and a step of measuring three-dimensional measurement data of the second space using the 3D scanner; (6) A computer-readable recording medium in which the program according to (4) or (5) is recorded. is provided. [Effects of the Invention]

[0009] According to the present invention, it is possible to efficiently survey spaces such as attics, underfloor spaces, and ceilings in traditional architecture such as shrines and temples. It is also possible to provide a method for generating composite point cloud data that can generate highly accurate composite point cloud data based on point cloud data obtained by surveying traditional architecture such as shrines and temples using a 3D scanner, a method for generating point cloud current status preservation drawings, a program, and a computer-readable recording medium. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram showing an outline of an example of traditional architecture such as a shrine or temple. [Figure 2] This is a flowchart showing the steps to use a 3D scanner to survey hidden spaces such as temples and shrines. [Figure 3] FIG. 1 is a block diagram showing a system configuration of a computer for generating composite point cloud data. [Figure 4] 10 is a flowchart showing a process for generating composite point cloud data. [Figure 5] FIG. 10 is a diagram showing an example of a point cloud current status saved drawing. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, a method for surveying a building and a method for generating composite point cloud data according to an embodiment of the present invention will be described with reference to the drawings. The method for surveying a building of the present invention is a method for surveying a second space that is partitioned as a separate space from a first space, which is an open space in traditional architecture, and is out of sight, and includes the steps of checking the status of an introduction path for a 3D scanner into the second space, introducing the 3D scanner into the second space if the introduction path does not impede the introduction of the 3D scanner, and acquiring three-dimensional survey data of the second space using the 3D scanner.

[0012] Furthermore, the method for generating composite point cloud data of the present invention includes the steps of acquiring first point cloud data obtained by surveying a first space, which is an open space in traditional architecture, with a 3D scanner, and second point cloud data obtained by surveying a second space, which is partitioned as a separate space from the first space in the traditional architecture and is not visible to the public, with a 3D scanner; creating joint groups by associating small groups, which are obtained by dividing a large space into a predetermined number of units and are set for each acquisition unit when acquiring the first point cloud data, with the second point cloud data at positions corresponding to the small groups; determining the position of the joint group in the large space; and generating composite point cloud data by combining the first point cloud data and the second point cloud data based on the determined position.

[0013] Fig. 1 is a diagram showing an outline of an example of traditional architecture such as a shrine or temple. As shown in Fig. 1, in a traditional wooden building 2, a foundation stone 6 is placed on the ground 4, and a pillar 8 is provided on the foundation stone 6. Furthermore, above the pillar 8, a girders 10 is provided as a horizontal structural member and joined to the pillar 8. A roof 11 is provided above the girders 10.

[0014] Additionally, ceiling material 12 is provided above the pillars 8 and below the girders 10 so as to be joined to the pillars 8, and floor material 14 is joined to the lower part of the pillars 8. Additionally, crossbeams 16 are joined to the pillars 8.

[0015] Each part is joined using traditional construction methods such as tenons and mortise joints, but metal fittings such as anchor bolts may also be used at the attachment points. The roofing material for the roof 11 is not particularly limited, but tiles, copper plates, thatch, etc. may be used. While the traditional architecture 2 has been described as having pillars 8 attached to cornerstones 6, it may also be constructed using foundations, etc.

[0016] Here, the open space formed between the floor material 14 and the ceiling material 12 is a space that can be used by users (for example, priests, head priests, worshippers, etc.) as a porch 17, veranda 18, outer sanctuary 20, and inner sanctuary 22.

[0017] On the other hand, the underfloor space 24 formed between the floor material 14 and the ground 4, the attic space 26 formed between the ceiling material 12 and the girders 10, and the attic space 28 formed between the girders 10 and the roof 11 are spaces that cannot be seen by users, that is, spaces that are hidden from view. Furthermore, the underfloor space 24, the attic space 26, and the attic space 28 are spaces that are not intended to be available to users.

[0018] Next, the procedure for measuring traditional architecture 2 such as a temple or shrine building using a 3D scanner will be described with reference to the flowchart shown in Figure 2. First, the exterior is measured using a 3D scanner (step S1), and then the interior of the traditional architecture 2, including the open spaces visible to users (porch 17, veranda 18, outer sanctuary 20, and inner sanctuary 22, etc.), is measured using a 3D scanner (step S2).

[0019] Next, a 3D scanner is used to survey the hidden spaces (underfloor space 24, attic space 26, and attic space 28). The hidden spaces (underfloor space 24, attic space 26, and attic space 28) have narrow entrances to enter these spaces (i.e., access points for bringing the 3D scanner in and out), and the interiors are narrow because they are not designed as spaces for users, and furthermore, they are not designed to let in light, so they tend to be dark.

[0020] First, it is confirmed whether or not the 3D scanner can be brought in through the entrance of a space that is not visible to the public (underfloor space 24, attic space 26, or attic space 28) (step S3). In other words, it is confirmed whether or not the opening is large enough to allow the 3D scanner to be brought in.

[0021] 3D scanners include contact and non-contact types, but it is preferable to use a non-contact type, i.e., a 3D laser scanner. The scanning method of the 3D laser scanner may be a pulse method or a phase difference method.

[0022] In addition, 3D scanners include handheld and stationary types, but stationary types are preferred from the viewpoint of measurement accuracy. In the present invention, for example, a Trimble X7 (external dimensions: width 178 mm × height 353 mm, depth 170 mm) can be used. Note that the Trimble X7 can perform surveying if the distance to the measurement object is at least 60 cm.

[0023] If the 3D scanner can be brought in through the entrance of a space that is not visible to the public (underfloor space 24, attic space 26, or attic space 28), the introduction route for the 3D scanner is confirmed (step S4). For example, it is checked whether the gap between pillars is large enough to allow the 3D scanner to pass through, whether the strength of the surface functioning as a floor can withstand the introduction of the 3D scanner, etc. If it is confirmed that there are no obstacles to the introduction of the 3D scanner, an introduction route that does not impede the introduction of the 3D scanner is determined.

[0024] For example, in the attic space 28, there may be many pillars for constructing the roof design, and the spacing between pillars and other structures may be narrow in the underfloor space 24. As a result, there are places between the structures where the 3D scanner cannot pass. Furthermore, there are places in the underfloor space 24 where a person must crawl to enter.

[0025] Furthermore, the surfaces that function as floors in the attic space 28 and ceiling space 26 are not intended for use by users, and so some areas are not strong enough. Furthermore, in traditional architecture with high cultural value, it is necessary to avoid damaging the building, especially when installing a 3D scanner. Therefore, when installing a 3D scanner, it is necessary to check whether there are any obstacles to its installation.

[0026] The distance that the 3D scanner can pass through and the strength of the surface that functions as the floor can be measured using a known measuring device, or a preliminary investigation can be carried out by a carpenter or other person familiar with the structure of traditional architecture such as shrine and temple buildings.

[0027] If it is determined that there is no problem with introducing the 3D scanner, the 3D scanner is introduced into an out-of-sight space (underfloor space 24, attic space 26 or attic space 28) based on the determined introduction route, and a survey of the out-of-sight space (underfloor space 24, attic space 26 or attic space 28) is performed using the 3D scanner (step S5).

[0028] The 3D scanner may be self-propelled, with a pre-programmed introduction path, and may be introduced along the pre-programmed introduction path, or may be remotely controlled. At the survey point, it is desirable to survey as many directions as possible, for example, to perform a 360° survey in multiple sessions. The survey may be performed remotely, for example. By following the above procedure, surveying of traditional architecture2 such as shrines and temples can be performed using a 3D scanner, and point cloud data can be obtained as 3D measurement data.

[0029] Next, we will explain the synthesis of point cloud data as three-dimensional measurement data obtained by surveying using a 3D scanner. Figure 3 is a block diagram showing the system configuration of a computer used to synthesize point cloud data. As shown in Figure 3, computer 30 includes a CPU 32, which is connected to memory 34, a communication unit 36 ​​that receives point cloud data from the 3D scanner, a storage unit 38 that stores point cloud data obtained by the 3D scanner, etc., a synthesized point cloud data generation unit 40 that synthesizes point cloud data using a program for generating synthesized point cloud data to generate synthesized point cloud data, a drawing output unit 42 that outputs various drawings based on the synthesized point cloud data, a display unit 44 configured, for example, by an LCD display that displays the drawings output by drawing output unit 42, and an input unit 46 for inputting various operation instructions to CPU 32.

[0030] 4 is a flowchart showing the point cloud data synthesis process. When a point cloud data synthesis instruction is input via the input unit 46, the CPU 32 reads the point cloud data from the storage unit 38 to the memory 34 (step S11). Here, the point cloud data stored in the storage unit 38 is data obtained by surveying the above-mentioned traditional architecture 2, such as a shrine or temple architecture, using a 3D scanner. The point cloud data is also obtained from the 3D scanner via the communication unit 36 ​​and stored in the storage unit 38, for example.

[0031] Next, the CPU 32 performs a synthesis process for the point cloud data. Generally, when surveying the same space using a 3D scanner in multiple directions or from multiple locations, setting a reference point for alignment allows for the generation of synthesized point cloud data with minimal error. For example, when synthesizing point cloud data obtained in the same space, the surveying direction of the 3D scanner is moved six times at intervals of approximately 60° from approximately the center of the space during the survey, thereby acquiring point cloud data for a total of 360°. In this case, a reference point can be set for each surveying direction. Then, the point cloud data obtained from one survey (for example, data of about 60°) can be defined as, for example, one large space, and the point cloud data obtained from the above six surveys, i.e., the point cloud data for the six large spaces, can be synthesized based on the reference point to obtain synthesized point cloud data.

[0032] In the present invention, as described above, it is necessary to synthesize point cloud data obtained by surveying the exterior, open spaces, and hidden spaces of traditional architecture 2 such as shrines and temples.

[0033] In traditional architecture 2 such as shrines and temples, the connection between the exterior and open spaces is visible, making it easy to set reference points for alignment. Therefore, when combining point cloud data obtained by surveying the exterior with point cloud data obtained by surveying the open spaces (porch 17, veranda 18, outer sanctuary 20, inner sanctuary 22, etc.), it is possible to obtain composite point cloud data with fewer errors based on the reference points.

[0034] On the other hand, when combining point cloud data obtained by surveying open spaces (such as the porch 17, veranda 18, outer sanctuary 20, and inner sanctuary 22) with point cloud data obtained by surveying hidden spaces (such as the underfloor space 24, attic space 26, and attic space 28), it is difficult to set a reference point for alignment, and even if it is possible to do so, combining the point cloud data as is tends to result in errors. In other words, as with the algorithm for combining point cloud data of large spaces as described above, using a reference point to combine the point cloud data will result in large errors. Therefore, in the present invention, small groups smaller than the large space are created, and joint groups are further created based on the small groups (step S12).

[0035] For example, if the point cloud data obtained in a single survey using a 3D scanner in an open space is defined as one large space, multiple small groups smaller than the large space are created. In other words, multiple small groups are created by dividing one large space into multiple parts. Then, for each small group position, a corresponding position in an out-of-sight space (an out-of-sight space located behind the open space) is combined to create a joint group. In other words, the joint group includes point cloud data of the open space included in the small group and point cloud data of the out-of-sight space at a position corresponding to the small group. Here, there is no particular limit to the number of small groups created in a large space, but it is possible to create several to several hundred small groups.

[0036] Regarding a joint group, for example, a joint group is a combination of point cloud data of a specified position on the ceiling obtained by surveying an open space and point cloud data of a corresponding position (behind the specified position on the ceiling) obtained by surveying the attic space 26.

[0037] In addition, a joint group is formed by combining point cloud data of a specified position on the floor obtained by surveying the open space with point cloud data of a corresponding position (the back side of the specified position on the floor) obtained by surveying the underfloor space 24.

[0038] In addition, when an open space and an attic space 28 are adjacent to each other, a joint group is formed by combining point cloud data of a specified position related to the main beam 10 obtained by surveying the open space and point cloud data of a corresponding position (the back side of the specified position related to the main beam 10) obtained by surveying the attic space 28.

[0039] Furthermore, if there is an attic space 26 between the open space and the attic space 28, the joint group may be a combination of point cloud data of a specified position on the ceiling obtained by surveying the open space and point cloud data of a corresponding position (behind the specified position on the ceiling) obtained by surveying the attic space 26 and the attic space 28, or a process may be performed to create the joint group twice.

[0040] When creating a joint group twice, a first joint group is created by combining point cloud data of a predetermined position related to attic space 28 obtained by surveying attic space 26 with point cloud data of a corresponding position (behind the predetermined position related to attic space 26) obtained by surveying attic space 28, and then a second joint group is created by combining point cloud data of a predetermined position corresponding to the first joint group obtained by surveying the open space with point cloud data included in the first joint group. In this case, the second joint group is used as the joint group in subsequent processing.

[0041] Next, the CPU 32 performs a process of making the joint group follow the large space (step S13), that is, a process of determining the position of the joint group in the large space. Then, the CPU 32 causes the composite point cloud data generating unit 40 to generate composite point cloud data based on the joint group (step S14).

[0042] This allows for the synthesis of point cloud data with fewer errors. In addition, the accuracy of the point cloud current status saved drawing (described later) is improved, allowing for the output of a drawing that is closer to the current status. Based on the obtained composite point cloud data, the CPU 32 causes the drawing output unit 42 to output a drawing in a predetermined format, and causes the display unit 44 to display the drawing.

[0043] For example, line drawings extracted from point cloud data can be generated as point cloud current status preservation drawings. An example of a point cloud current status preservation drawing is shown in Figure 5. Point cloud current status preservation drawings can be positioned as faithful preservation drawings that preserve the current state. Point cloud current status preservation drawings can also be used to archive traditional architecture with high cultural value, such as historical buildings, or as digital twins. In addition, after synthesizing the point cloud data, it can be converted into two dimensions, dimensions can be added, and construction drawings and marking out drawings can be generated. Furthermore, the point cloud current status preservation drawings, construction drawings, marking out drawings, etc. can be displayed not only on the display unit 44 but also printed out on paper for use.

[0044] The computer 2 may be a tablet terminal, and if the operator determines that the drawing displayed on the display unit 44 based on the surveying results does not fully display the current situation, the process of generating the composite point cloud data may be re-executed.Furthermore, since the accuracy of the point cloud data obtained may be affected by the measurement angle, installation position, etc. of the 3D scanner, the survey may be performed again using the 3D scanner.

[0045] In addition to the above-mentioned processing, if necessary, during or after synthesis, it is possible to perform trimming of unnecessary parts as needed, and after synthesizing the point cloud data, it is possible to perform processing such as filling in data or smoothing for parts where no point cloud data exists (i.e., parts where survey data could not be obtained using a 3D scanner).

[0046] In the above-described embodiment, first point cloud data obtained by surveying a first space, which is an open space in traditional architecture, with a 3D scanner, and second point cloud data obtained by surveying a second space, which is partitioned as a separate space from the first space in the traditional architecture and is out of sight, with a 3D scanner, are acquired, and small groups obtained by dividing a large space into a predetermined number of units, which are set for each acquisition unit when acquiring the first point cloud data, are combined with the second point cloud data at positions corresponding to the small groups to create joint groups, and the position of the joint group in the large space is determined. A program for generating composite point cloud data by combining the first point cloud data and the second point cloud data based on the determined position can be downloaded via a network such as the Internet and incorporated into a computer, thereby causing the computer to function so as to perform the above-described composite point cloud data generation process.

[0047] The program may be recorded on a computer-readable recording medium such as a flexible disk, a CD-ROM, a DVD, a Blu-ray, etc. That is, the program may be read from the computer-readable recording medium and installed in a computer, causing the computer to function so as to perform the above-described process of generating composite point cloud data. [Explanation of symbols]

[0048] 2...Traditional architecture such as shrines and temples, 6...Foundation stone, 10...Beam, 11...Roof, 12...Ceiling material, 14...Floor material, 17...Veranda, 18...Wide veranda, 20...Outer sanctuary, 22...Inner sanctuary, 24...Underfloor space, 26...Attic space, 28...Attic space

Claims

1. A step of acquiring first point cloud data obtained by surveying a first space, which is an open space in a traditional building, with a 3D scanner, and second point cloud data obtained by surveying a second space, which is partitioned as a space separate from the first space in the traditional building and is out of sight, with a 3D scanner; creating a joint group by associating small groups obtained by dividing a large space set for each acquisition unit when acquiring the first point cloud data into a predetermined number of units with the second point cloud data at positions corresponding to the small groups; determining a position of the joint group in the larger space; generating composite point cloud data by combining the first point cloud data and the second point cloud data based on the determined position; A method for generating synthetic point cloud data, comprising:

2. The second point cloud data is A step of confirming the load capacity of an introduction path of the 3D scanner into the second space; When the load capacity of the introduction path does not interfere with the introduction of the 3D scanner, introducing the 3D scanner into the second space; measuring three-dimensional measurement data of the second space by the 3D scanner; The synthetic point cloud data generating method according to claim 1 , wherein the data is obtained by

3. A method for generating a point cloud as-is preserved drawing, comprising the step of extracting a line drawing based on the composite point cloud data according to claim 1 or 2.

4. Computer A means for acquiring first point cloud data obtained by surveying a first space, which is an open space in a traditional building, with a 3D scanner, and second point cloud data obtained by surveying a second space, which is partitioned as a space separate from the first space in the traditional building and is not visible to the public, with a 3D scanner; a means for combining small groups obtained by dividing a large space set for each acquisition unit when acquiring the first point cloud data into a predetermined number of units with the second point cloud data at positions corresponding to the small groups to create a joint group; means for determining the position of said joint group in said large space; means for generating composite point cloud data by combining the first point cloud data and the second point cloud data based on the determined position; A program to function as a

5. The second point cloud data is A step of confirming the load capacity of an introduction path of the 3D scanner into the second space; When the load capacity of the introduction path does not interfere with the introduction of the 3D scanner, introducing the 3D scanner into the second space; measuring three-dimensional measurement data of the second space by the 3D scanner; 5. The program according to claim 4, wherein the data is obtained by

6. 6. A computer-readable recording medium having the program according to claim 4 or 5 recorded thereon.

Images