Registration device, registration method, and program

Abstract

This registration device (1) comprises: a jig (2) provided with three or more spheres that are linked, the jig (2) being disposed within a range in which measurement is possible from each of a plurality of measurement positions inside a structure (20); a measurement unit (11) for using a three-dimensional laser scanner to obtain, from each of the plurality of measurement positions, three-dimensional point cloud data regarding the structure (20) including, in the measurement range, the entirety of the jig (2) disposed at the same position; a first computation unit (12) for calculating, from the three-dimensional point cloud data, the center coordinates of the three or more spheres provided to the jig (2), and generating initial position alignment data with regards to initial position alignment between the plurality of items of three-dimensional point cloud data such that the center coordinates coincide; and a second computation unit (13) for generating final position alignment data for final position alignment between the plurality of items of three-dimensional point cloud data, using a position alignment algorithm, from the initial position alignment data.

Description

[Technical Field]

[0001] The present invention relates to a registration device, a registration method, and a program for aligning multiple pieces of three-dimensional point cloud data measured from different measurement points. [Background technology]

[0002] Conventionally, to prevent accidents such as the collapse or collapse of the lining surface of a structure, deformations such as lifting, peeling, and uneven deformation on the lining surface of the structure are measured. At this time, measurements are performed using a 3D laser scanner, and 3D point cloud data is acquired. The interior cross section of the structure is then accurately measured from the acquired 3D point cloud data. However, if the measurement area is large or if blind spots due to installed objects exist, it is not possible to acquire 3D point cloud data for the entire object in a single measurement. For this reason, multiple 3D point cloud data measured from different measurement points are acquired. When acquiring 3D point cloud data from multiple measurement points, registration is required to align each 3D point cloud data to the same coordinate system. In this disclosure, registration refers to the alignment of 3D point cloud data acquired from multiple measurement points.

[0003] Non-Patent Document 1 describes a technique for selecting highly accurate point cloud data in multi-point measurements using a laser scanner. The technique described in Non-Patent Document 1 involves placing multiple targets within the measurement range, extracting targets from 3D point cloud data, and then associating the point cloud data with each other via the corresponding targets, thereby performing registration (alignment) between the 3D point cloud data.

[0004] Non-Patent Document 2 describes a technique for efficiently acquiring point cloud data by combining a terrestrial laser scanner (TLS) and a satellite positioning system (GNSS (Global Navigation Satellite System)). The technique described in Non-Patent Document 2 performs registration (alignment) of 3D point cloud data by clarifying and measuring the measurement positions of measuring instruments such as a 3D laser scanner and a target. [Prior art documents] [Non-patent literature]

[0005] [Non-Patent Document 1] Jun Sakurai and five others, "Development of High-Precision Point Cloud Data Selection Technique for Multi-Point Measurement Using Laser Scanners," Journal of the Japan Society of Civil Engineers, F3 (Civil Engineering Informatics), Vol. 72, No. 2, I_209-I_218, 2016. [Non-patent document 2] Kazushi Moriishi and 1 other collaborators, "Improving the efficiency of point cloud data acquisition and expanding the scope of application in ICT paving construction," Journal of the Japan Society of Civil Engineers, Vol. 75, No. 2, I_77-I_85, 2019. Summary of the Invention [Problem to be solved by the invention]

[0006] However, the technology described in Non-Patent Document 1 requires the installation of multiple targets. Therefore, when the measurement range becomes wider, the number of targets needs to be increased according to the number of measurements, and the installation of the targets takes time. Furthermore, when the working space is limited, it is difficult to find a place to install multiple targets.

[0007] On the other hand, the technology of Non-Patent Document 2 requires expensive measuring equipment equipped with a satellite positioning system (GNSS (Global Navigation Satellite System)) or the like. In addition, GNSS cannot accurately determine the measurement position in structures with limited communication environments, such as underground tunnels, and therefore cannot perform registration. As such, the technology for acquiring point cloud data using GNSS is suitable for measurements in places with good communication environments, but has the problem that it is not suitable for capturing deformations that occur in places with limited communication environments. [Means for solving the problem]

[0008] In view of the above circumstances, the object of the present invention is to provide a registration device, a registration method, and a program that aligns multiple 3D point cloud data obtained from different measurement points based on the 3D coordinates of a jig having three or more connected spheres installed within the measurement range.

[0009] In order to solve the above problem, the registration device of the first embodiment is a registration device that aligns multiple 3D point cloud data, and includes: a jig having three or more connected spheres arranged within a range measurable from each of multiple measurement points inside a structure; a measurement unit that uses a 3D laser scanner to acquire 3D point cloud data of the structure from each of the multiple measurement points, the measurement range of which includes the jig arranged in the same position; a first calculation unit that calculates the center coordinates of the three or more spheres arranged on the jig from the 3D point cloud data and generates initial alignment data between the multiple 3D point cloud data so that the center coordinates match; and a second calculation unit that uses an alignment algorithm to generate final alignment data between the multiple 3D point cloud data from the initial alignment data.

[0010] In order to solve the above problem, a registration method according to a first embodiment is a registration method for aligning multiple pieces of 3D point cloud data, the registration method including the steps of: a measurer placing a jig having three or more connected spheres in a range where the jig can be measured from each of multiple measurement points inside a structure; a registration device using a 3D laser scanner to acquire 3D point cloud data of the structure from each of the multiple measurement points, the measurement range including the entire jig placed at the same position; a registration device calculating, from the 3D point cloud data, central coordinates of the three or more spheres included in the jig; a registration device generating, from the initial alignment data, initial alignment data between the multiple pieces of 3D point cloud data so that the central coordinates match; and a registration device generating, from the initial alignment data, final alignment data between the multiple pieces of 3D point cloud data using an alignment algorithm.

[0011] In order to solve the above problem, a program according to a first embodiment causes a computer to function as the registration device. [Effects of the Invention]

[0012] According to the present disclosure, when performing measurements using a 3D laser scanner on structures in a non-GNSS environment, the time required for setting up targets can be simplified while also enabling efficient registration of 3D point cloud data. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is a block diagram showing an example of the configuration of a registration device according to a first embodiment. [Figure 2] 1A to 1C are a plan view, a front view, and a side view showing an example of the configuration of a jig made up of three connected spheres. [Figure 3] FIG. 1 is a diagram illustrating an example of measurement in a structure. [Figure 4] 5A to 5C are diagrams illustrating the process of aligning coordinate systems by a first calculation unit according to the first embodiment. [Figure 5] 4 is a flowchart showing an example of a registration method executed by the registration device according to the first embodiment. [Figure 6] FIG. 10 is a block diagram illustrating an example of the configuration of a registration device according to a second embodiment. [Figure 7] FIG. 10 is a diagram illustrating a plane element set in the second embodiment. [Figure 8] 10A to 10C are diagrams illustrating a process of aligning a coordinate system by a first calculation unit according to the second embodiment. [Figure 9] FIG. 2 is a block diagram showing a schematic configuration of a computer that functions as a registration device. DETAILED DESCRIPTION OF THE INVENTION

[0014] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. The present invention is not limited to the following embodiments, and various modifications can be made within the scope of the present invention.

[0015] (First embodiment) Fig. 1 is a block diagram showing an example of the configuration of a registration device 1 according to the first embodiment. As shown in Fig. 1, the registration device 1 includes a measurement unit 11, a first calculation unit 12, and a second calculation unit 13. The registration device 1 performs measurements from each of a plurality of measurement points using a 3D laser scanner, acquires 3D point cloud data from each of the plurality of measurement points, and then aligns the plurality of 3D point cloud data.

[0016] The measurement unit 11, the first calculation unit 12, and the second calculation unit 13 constitute a control calculation circuit (controller) 30. The control calculation circuit 30 may be constituted by dedicated hardware such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field-Programmable Gate Array), or may be constituted by a processor, or may be constituted by including both.

[0017] The jig 2 is placed by a measurement worker within a range where the jig 2 can be measured from each of multiple measurement points inside the structure 20. The jig 2 comprises three or more connected spheres. In the embodiment described below, the jig 2 comprises three connected spheres. When performing measurement using a 3D laser scanner, one jig 2 is placed within the measurement range in advance. Figure 2 is a plan view, a front view, and a side view showing an example configuration of the jig 2 composed of three connected spheres. As shown in the area a indicated by the dashed line in the front view of Figure 2, the jig 2 is composed of three spheres arranged to form a single plane. The reason for using spheres is that they can be measured as an object of the same shape from multiple different measurement points.

[0018] 3 is a diagram showing an example of measurement in a structure 20 (underground tunnel). As shown in Fig. 3, a jig 2 having three connected spheres is placed so that the entire outer shape of the jig 2 (the three spheres) falls within the measurement range from measurement point A and measurement point B.

[0019] The measurement unit 11 uses a 3D laser scanner to acquire 3D point cloud data of the structure 20 from each of a plurality of different measurement points, the measurement range of which includes the entire jig 2 (three or more spheres) that has three or more spheres and that has been installed in advance at the same position inside the structure. The measurement unit 11 outputs the 3D point cloud data of the structure 20 acquired from each measurement point to the first calculation unit 12.

[0020] The three-dimensional point cloud data to be registered must include point cloud data of the jig 2 installed at the same position, measured from each of a plurality of different measurement points.

[0021] The first calculation unit 12 calculates the central coordinates of three or more spheres provided on the jig 2 from the 3D point cloud data acquired at each of the multiple measurement points, and generates initial alignment data that completes initial alignment between the multiple 3D point cloud data so that the central coordinates match. The first calculation unit 12 outputs the initial alignment data to the second calculation unit.

[0022] FIG. 4 is a diagram illustrating the process of aligning coordinate systems performed by the first calculation unit 12 according to the first embodiment. As shown in FIG. 4, (i) the first calculation unit 12 reads three-dimensional point cloud data (referred to as data A to X) measured at each of a plurality of measurement points A to X. The data reading is repeated according to the number of pieces of data to be registered. (ii) Next, for each piece of data A to X, the first calculation unit 12 extracts three spheres from the three-dimensional point cloud data to be registered and calculates the center coordinates of each sphere. (iii) Then, the first calculation unit 12 aligns the coordinate systems so that the center coordinates of the three spheres of each piece of data A to X coincide (so that planes formed by the center coordinates of the three spheres overlap), performs initial alignment between the plurality of pieces of three-dimensional point cloud data A to X, and generates initially aligned data A1 to X1.

[0023] The second calculation unit 13 generates final registration data A2 to X2 between multiple pieces of 3D point cloud data using a registration algorithm from the initial registration data generated by the first calculation unit 12. After the first calculation unit 12 performs initial registration, the second calculation unit 13 generates final registration data A2 to X2 using a registration algorithm for 3D point cloud data such as ICP (Iterative Closest Point). The first calculation unit performs initial registration using only the center coordinates of the sphere, and the second calculation unit performs final registration using all of the 3D point cloud data. The reason for performing registration in two stages is that performing registration by wide-range search using ICP from the beginning may result in a localized solution, and therefore the present disclosure employs a method of applying ICP after performing initial registration.

[0024] FIG. 5 is a flowchart showing an example of a registration method executed by the registration device 1 according to the first embodiment.

[0025] In step S101, a measurer places one jig 2 having three connected spheres within a range where the jig 2 can be measured from each measurement point.

[0026] In step S102, the measurement unit 11 uses a three-dimensional laser scanner to repeatedly measure the structure 20 from each of the multiple measurement points A to X, with the measurement range including the entire jig 2 (three spheres) placed at the same position, and obtains three-dimensional point cloud data A to X of the structure 20.

[0027] In step S103, the first calculation unit 12 reads the three-dimensional point cloud data A to X of the structure 20.

[0028] In step S104, the first calculation unit 12 calculates the coordinates of the centers of the three spheres provided on the jig 2 from the three-dimensional point cloud data A to X.

[0029] In step S105, the first calculation unit 12 generates initial alignment data between multiple three-dimensional point cloud data from the central coordinates of the three spheres calculated for each of the three-dimensional point cloud data A to X so that the central coordinates of the three spheres match.

[0030] In step S106, the second calculation unit 13 generates final registration data A2 to X2 between the plurality of three-dimensional point cloud data from the initial registration data using a registration algorithm.

[0031] If adding measurement points without changing the range for which you want to acquire 3D point cloud data, place jig 2 in the same position and perform measurements at the added measurement points according to the flowchart in Fig. 5. If you want to change the range for which you want to acquire 3D point cloud data, move the position of jig 2, set the first measurement point as measurement point A, and execute the flowchart in Fig. 5 from the beginning.

[0032] According to the registration device 1 of this embodiment, when performing measurements using a 3D laser scanner on structures in a non-GNSS environment, the time required for setting up targets can be simplified while at the same time enabling efficient registration of 3D point cloud data.

[0033] (Second embodiment) FIG. 6 is a block diagram showing an example of the configuration of a registration device 1′ according to the second embodiment. As shown in FIG. 6, the registration device 1′ includes a measurement unit 11, a first calculation unit 12′, and a second calculation unit 13. The registration device 1′ aligns multiple pieces of 3D point cloud data. The registration device 1′ according to this embodiment differs from the registration device 1 according to the first embodiment in part of the processing function of the first calculation unit 12′. The same components as those in the first embodiment are denoted by the same reference numerals as those in the first embodiment, and descriptions thereof will be omitted where appropriate.

[0034] A control and arithmetic circuit (controller) 30′ is configured by the measurement unit 11, the first arithmetic unit 12′, and the second arithmetic unit 13. The control and arithmetic circuit 30′ may be configured by dedicated hardware such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field-Programmable Gate Array), or may be configured by a processor, or may be configured by including both.

[0035] The first calculation unit 12′ calculates the center coordinates of three or more spheres, and then generates initial alignment data between multiple 3D point cloud data so that the coordinates of a set of plane elements, which are a set of multiple points that make up a polygonal plane with the center coordinates as vertices, match.

[0036] Fig. 7 is a diagram illustrating a plane element set in the second embodiment. When the jig 2 includes three connected spheres, the central coordinates of the three spheres are expressed as (xa1, ya1, za1), (xa2, ya2, za2), and (xa3, ya3, za3), as shown in Fig. 7. In contrast, the plane element set (Pa0, Pa1, ..., Pan) is not the three points of the central coordinates of the three spheres, but is a set of multiple points that form a plane, as shown in the balloon in Fig. 7.

[0037] In order to match the coordinate systems of the two sets of plane elements, the following formula (1) is used to find the rotation matrix R and the parallel matrix T so that the square error between the sets of plane elements is minimized.

number

[0038] Furthermore, when there are three or more plane element sets (Pc, etc.), the above formula (1) is applied based on the coordinate system calculated using Pa and Pb.

[0039] FIG. 8 is a diagram illustrating the process of aligning coordinate systems performed by a first calculation unit 12′ according to the second embodiment. As shown in FIG. 8, (i) the first calculation unit 12′ reads three-dimensional point cloud data (referred to as data A to X) measured at each of a plurality of measurement points A to X. The data reading is repeated depending on the number of pieces of data to be registered. (ii) Next, for each piece of data A to X, the first calculation unit 12′ extracts three spheres from the three-dimensional point cloud data to be registered and calculates the center coordinates of each sphere. (iii) Furthermore, the first calculation unit 12′ calculates plane element sets Pa to Px enclosed by the center coordinates of the three spheres. (iv) Finally, the first calculation unit 12′ aligns the coordinate systems so that the center coordinates of the three spheres or the coordinates of the plane element sets of each piece of data A to X match, thereby performing initial alignment between the plurality of pieces of three-dimensional point cloud data A to X and generating initially aligned data A1′ to X1′.

[0040] According to the registration device 1' of this embodiment, by using a set of plane elements, which is a set of multiple points that make up a plane, it is expected that the processing speed will be increased toward convergence when setting a coordinate system.

[0041] A computer capable of executing program instructions can also be used to cause the above-described registration apparatuses 1 and 1' to function. FIG. 9 is a block diagram showing a schematic configuration of a computer functioning as the registration apparatuses 1 and 1'. Here, the computer functioning as the registration apparatuses 1 and 1' may be a general-purpose computer, a dedicated computer, a workstation, a PC (Personal Computer), an electronic notepad, etc. The program instructions may be program code, code segments, etc. for performing necessary tasks.

[0042] 9, the computer 100 includes a processor 110, a memory unit including a ROM (Read Only Memory) 120, a RAM (Random Access Memory) 130, and a storage 140, an input unit 150, an output unit 160, and a communication interface (I / F) 170. Each component is connected to each other via a bus 180 so as to be able to communicate with each other.

[0043] The ROM 120 stores various programs and various data. The RAM 130 temporarily stores programs or data as a working area. The storage 140 is configured with an HDD (Hard Disk Drive) or an SSD (Solid State Drive) and stores various programs including an operating system and various data. In the present disclosure, the programs related to the present disclosure are stored in the ROM 120 or the storage 140.

[0044] Specifically, the processor 110 is a CPU (Central Processing Unit), MPU (Micro Processing Unit), GPU (Graphics Processing Unit), DSP (Digital Signal Processor), SoC (System on a Chip), etc., and may be configured with multiple processors of the same or different types. The processor 110 reads a program from the ROM 120 or storage 140 and executes the program using the RAM 130 as a working area, thereby controlling the above components and performing various arithmetic processing. Note that at least a part of these processing contents may be realized by hardware.

[0045] The program may be recorded on a recording medium readable by the registration devices 1 and 1'. If such a recording medium is used, the program can be installed in the registration devices 1 and 1'. Here, the recording medium on which the program is recorded may be a non-transitory recording medium. The non-transitory recording medium is not particularly limited, and may be, for example, a CD-ROM, a DVD-ROM, a USB (Universal Serial Bus) memory, or the like. The program may also be downloaded from an external device via a network.

[0046] The following additional notes are provided regarding the above-described embodiments.

[0047] (Additional note 1) A registration device for aligning a plurality of 3D point cloud data, a jig having three or more connected spheres arranged within a range measurable from each of a plurality of measurement points inside the structure; a controller that uses a 3D laser scanner to acquire 3D point cloud data of the structure from each of the multiple measurement points, the measurement range of which includes the entire jig placed at the same position; calculates the center coordinates of the three or more spheres that the jig has from the 3D point cloud data; generates initial alignment data between the multiple 3D point cloud data so that the center coordinates match; and generates final alignment data between the multiple 3D point cloud data from the initial alignment data using an alignment algorithm. (Additional note 2) The registration device described in Appendix 1, wherein the controller calculates the center coordinates of the three or more spheres, and then generates initial alignment data between the plurality of 3D point cloud data so that the coordinates of a set of plane elements, which is a set of a plurality of points that constitute a polygonal plane with the center coordinates as vertices, match. (Additional note 3) A registration method for aligning a plurality of 3D point cloud data, comprising: A measurement technician places a jig having three or more connected spheres within a range where the jig can be measured from each of a plurality of measurement points inside the structure; A registration method in which a registration device uses a 3D laser scanner to acquire 3D point cloud data of the structure, with the measurement range including the entire jig placed at the same position, from each of the multiple measurement points, calculates the center coordinates of the three or more spheres that the jig has from the 3D point cloud data, generates initial alignment data between the multiple 3D point cloud data so that the center coordinates match, and generates final alignment data between the multiple 3D point cloud data from the initial alignment data using an alignment algorithm. (Additional note 4) A non-transitory storage medium storing a program executable by a computer, the non-transitory storage medium storing the program causing the computer to function as the registration device described in appendix 1 or 2.

[0048] Although the above-described embodiments have been described as typical examples, it will be apparent to those skilled in the art that many modifications and substitutions can be made within the spirit and scope of the present disclosure. Therefore, the present invention should not be interpreted as being limited by the above-described embodiments, and various modifications or alterations are possible without departing from the scope of the claims. For example, multiple building blocks shown in the block diagrams of the embodiments can be combined into one, or one building block can be divided. [Explanation of symbols]

[0049] 1,1' Registration Device 2. Jig (a jig with three or more connected spheres) 11 Measurement unit (3D laser scanner) 12,12′ 1st ​​calculation section 13 Second calculation section 20 Structures 30, 30' Control operation circuit (controller) 100 computers 110 processors 120 ROM 130 RAM 140 Storage 150 Input section 160 Output section 170 Communication Interface (I / F) 180 Bus

Claims

1. A registration device for aligning a plurality of three-dimensional point cloud data, comprising: a jig having three or more connected spheres arranged within a range measurable from each of a plurality of measurement points inside the structure; a measurement unit that uses a three-dimensional laser scanner to acquire three-dimensional point cloud data of the structure from each of the plurality of measurement points, the measurement range of which includes the entire jig placed at the same position; a first calculation unit that calculates center coordinates of the three or more spheres included in the jig from the three-dimensional point cloud data, and generates initial alignment data between the plurality of three-dimensional point cloud data so that coordinates of a set of plane elements that are a set of a plurality of points constituting a polygonal plane with the center coordinates as vertices and are surrounded by the center coordinates match; a second calculation unit that generates final registration data between the plurality of three-dimensional point cloud data from the initial registration data using a registration algorithm; A registration device comprising:

2. A registration method for aligning a plurality of three-dimensional point cloud data, comprising: a step of placing a jig having three or more connected spheres by a measurement person within a range where the entire jig can be measured from each of a plurality of measurement points within the structure; a step of acquiring, by a registration device using a three-dimensional laser scanner, three-dimensional point cloud data of the structure from each of the plurality of measurement points, the measurement range of which includes the entire jig arranged at the same position; calculating, by the registration device, center coordinates of the three or more spheres provided on the jig from the three-dimensional point cloud data; generating, by the registration device, initial registration data between the plurality of three-dimensional point cloud data so that coordinates of a set of plane elements, which are a set of a plurality of points constituting a polygonal plane with the center coordinates as vertices and are surrounded by the center coordinates, match; generating final registration data between the plurality of 3D point cloud data from the initial registration data using a registration algorithm by the registration device; A registration method comprising:

3. A program for causing a computer to function as the registration device according to claim 1.

Images