Construction gauge measurement method and program

Abstract

To obtain an effective measurement result to an operator while utilizing three-dimensional point group data in construction gauge measurement with a building or the like as an object.SOLUTION: Rail track data RM is determined from three-dimensional point group data. Vehicle type two-dimensional cross sectional models m2 obtained by changing a contour shape by considering a cant, a transition curve, or the like are arranged on a rail track RM. A continuous external surface connecting the vehicle type two-dimensional cross sectional models m2 obtained by changing the contour shape is formed by graphic processing to generate a virtual vehicle type 3D model M1. Then, a separation distance between a construction gauge space area N1 corresponding to the vehicle type 3D model and a station PF (an edge part of a roof PM, in particular) is calculated.SELECTED DRAWING: Figure 9

Description

[Technical Field]

[0001] The present invention relates to a construction gauge established for vehicles such as railways, and in particular to measuring the distance between the construction gauge and its surrounding objects (hereinafter referred to as construction gauge measurement). [Background technology]

[0002] To ensure safe passage for trains that run along tracks and automobiles that run on roads, a "construction gauge" is established, which indicates the spatial range within which buildings and structures must not be erected. For example, in the case of railways, it is necessary to measure the distance between the construction gauge and buildings such as stations that are erected along the tracks, railway equipment such as signals and utility poles, as well as obstacles (such as trees) that exist around the tracks, and to confirm whether they are within the construction gauge.

[0003] In recent years, attempts have been made to utilize 3D point cloud data for measuring the construction limits of buildings and railway facilities around railway tracks, and for detecting obstacles such as trees. 3D point cloud data is 3D information acquired by 3D measurement of an object using a stereo method (active stereo method, passive stereo method), ToF (Time-of-Flight) method, etc., and each point in the point cloud data has 3D coordinate values.

[0004] The system then measures the shortest distance between a measurement point defined in the 3D point cloud data and the construction gauge area, and determines whether or not the point cloud data exists within the construction gauge area. This allows for confirmation that railway equipment and other objects exist outside the construction gauge area, or for detection of obstacles such as trees within the construction gauge area. The measured (calculated) distance from the construction gauge can be displayed numerically, or it can be visually expressed using a visualized image with coloring, etc.

[0005] Patent Documents 1 to 7 listed below describe methods for measuring construction gauges and detecting obstacles using three-dimensional point cloud data.

[0006] The data analysis device described in Patent Document 1 runs a data collection vehicle equipped with a three-dimensional measuring device along a railway line (hereinafter also referred to as a track or a rail) to acquire three-dimensional point cloud data and determine whether or not an obstacle exists within the construction gauge. Specifically, from the three-dimensional point cloud data acquired by running on a straight section, two-dimensional point cloud data of a cross section perpendicular to the running direction is extracted at predetermined distance intervals, a construction gauge area based on the running vehicle is set in accordance with the point cloud data of the railway line identified from the two-dimensional point cloud data of the cross section, and obstacles are detected by determining whether or not point cloud data exists within the construction gauge area.

[0007] In the construction gauge measurement drawing creation device described in Patent Document 2, as in Patent Document 1, a data collection vehicle equipped with a 3D measuring device is driven along the track to acquire 3D point cloud data. A 2D plane perpendicular to the track is then defined, and the shortest distance between the measurement point and the construction gauge (the distance to the closest point) is calculated. Measurement points that are within the construction gauge or nearby measurement points are colored or shaded according to the distance, and an image visualizing the measurement results is displayed.

[0008] As described in Patent Documents 1 and 2, 3D point cloud data of a measurement target is acquired by mounting a laser scanner (LIDAR) on a railway vehicle or maintenance vehicle and collecting the 3D point cloud data while the vehicle is traveling. Various methods have been proposed for extracting rail point cloud data from the 3D point cloud data acquired using such a traveling vehicle. For example, the vehicle-like correction device described in Patent Document 3 extracts point cloud data of the outer portion of the rail and the trackbed portion including the sleepers.

[0009] Also, in the track transportation system described in Patent Document 4, a 3D LIDAR (Laser Imaging Detection and Ranging) is installed on a running vehicle, and 3D point cloud data around the vehicle is acquired at multiple positions on the track. Then, matching is performed with a rail shape model stored in a database, and the point cloud data of the rail is identified and extracted from the 3D point cloud data.

[0010] The method for detecting obstacles in construction gauge measurement disclosed in Patent Document 1 is configured to detect obstacles based on a cross section perpendicular to the direction of vehicle travel (railway). Therefore, when detecting obstacles over the entire vehicle travel section, it takes a huge amount of time for calculation.

[0011] Furthermore, in Patent Document 1, it is assumed that the track is straight, and the structure gauge measurement is performed based on point cloud data on a cross section perpendicular to the track. However, many of the tracks used by running vehicles have complex trajectories that include not only straight sections, but also straight and curved sections. Furthermore, the structure gauge area set in curved sections is an area that differs from the structure gauge area in straight sections.

[0012] Specifically, to prevent rolling stock from tipping outward due to centrifugal force on curved sections of the track, the outer rail of the track is made higher than the inner rail, and a cant (incline) is set. Furthermore, a transition curve is set between the straight and curved sections, creating a complex track trajectory (route) with a gradually decreasing curvature over its entire length. The cant is determined according to this complex track trajectory, and the construction gauge area must also be set in an area (specifically, a rotated area) that is different from the straight sections.

[0013] Furthermore, because trains traveling on curved tracks tend to be biased, the construction gauge area is expanded accordingly to check that this does not cause any problems for the trains traveling on them.As the construction gauge area differs depending on the straight and curved sections of the track, or the radius and cant of the curved sections, construction gauge measurements that utilize 3D point cloud data must also take into account the track trajectory (route).

[0014] The structure gauge point cloud determination system described in Patent Document 5 defines a structure gauge area for straight and curved sections and detects obstacles within the structure gauge area. Specifically, for each fixed section of the curve, the structure gauge frame is varied and expanded according to the cant and curve radius, and this fixed section is defined as a structure gauge frame box. A two-dimensional plane is then defined for each box, and the variable structure gauge frame is projected onto the two-dimensional plane. Obstacles within the structure gauge frame are detected by determining whether point cloud data exists within the variable structure gauge frame projected onto the two-dimensional plane, and the point cloud data of the obstacles is color-coded and displayed on the screen.

[0015] Furthermore, Patent Document 6 describes a method for a structure gauge measurement device that connects two-dimensional structure gauge regions to generate a three-dimensional structure gauge region. Specifically, first, multiple two-dimensional structure gauge regions are arranged along track center data. Then, each structure gauge region is modified based on the curve radius and cant amount at that position, as well as the rail width expansion (slack) provided for vehicle running stability in curved sections. A connected region is generated as a three-dimensional polygon region for each modified structure gauge region, and it is determined whether point cloud data exists within the three-dimensional polygon region.

[0016] In this way, Patent Documents 5 and 6 are configured to automatically extract point cloud data that represent obstacles within a three-dimensional construction gauge area. On the other hand, a method has also been proposed in which, instead of automatically detecting obstacles through construction gauge measurement, construction gauge measurement is performed at measurement points determined in response to input operations by an operator.

[0017] The construction gauge display device described in Patent Document 7 performs three-dimensional coordinate transformation (specifically, rotation) of the construction gauge frame based on information such as the rail position, the difference in elevation between the left and right rails, and the rail extension direction, and displays the two-dimensional construction gauge frame in three dimensions, aligned with the points specified by the operator. [Prior art documents] [Patent documents]

[0018] [Patent Document 1] Japanese Patent Application Publication No. 2010-202017 [Patent Document 2] International Publication No. 2015 / 198423 [Patent Document 3] Japanese Patent Publication No. 2020-132094 [Patent Document 4] Japanese Patent Publication No. 2022-71407 [Patent Document 5] Japanese Patent Application Publication No. 2017-19388 [Patent Document 6] Patent Publication No. 2021-11240 [Patent Document 7] Japanese Patent Application Laid-Open No. 2017-165133 Summary of the Invention [Problem to be solved by the invention]

[0019] The targets of construction gauge measurements along railway tracks include buildings such as stations (platforms). Buildings such as stations are constructed along straight or curved sections of the tracks, and their length is designed to allow passengers to board and disembark trains with various formations (number of cars). Therefore, construction gauge measurements must be carried out over a wide area of ​​the building.

[0020] For operators involved in measuring the construction gauge of buildings, it is important not to simply determine whether or not there is an obstacle on the tracks (within the construction gauge area), but to understand how far the separation distance from a moving vehicle changes depending on the location of the building, and to confirm where the closest point of approach for a moving vehicle is and whether the separation distance is secured there. This is the same for measuring the construction gauge of structures other than buildings, and for measuring the construction gauge of vehicles such as automobiles traveling on roads.

[0021] Therefore, when measuring the construction gauge limits of buildings and other structures, it is necessary to obtain measurement results that are useful to the operator while using 3D point cloud data. [Means for solving the problem]

[0022] The construction gauge measurement method of the present invention is executed by a computer, a signal processing circuit, a processor, etc., and can be realized by a physical configuration capable of handling the vast amount of data known as 3D point cloud data. One aspect of the construction gauge measurement method of the present invention determines data on the railway or road (hereinafter referred to as the railway, etc.) from 3D point cloud data obtained by 3D measurement of the railway or road and buildings or structures (hereinafter referred to as buildings, etc.) located along the railway, etc., places a 3D construction gauge model of a vehicle having a length corresponding to the length of all or part of the building, etc., on the railway, etc. data having a trajectory including curved sections, and calculates the distance between the building, etc. and the boundary surface of the construction gauge space region determined according to the 3D construction gauge model placed on the railway, etc. data, based on the 3D point cloud data. Then, a 3D construction gauge model having a curved outer surface formed in accordance with the curvature of the railway, etc. data is placed on the railway, etc. data.

[0023] Another aspect of the present invention, a construction limit measurement method, determines data for the railways, etc. from three-dimensional point cloud data obtained by three-dimensionally measuring railways or roads (hereinafter referred to as railways, etc.) and buildings or structures (hereinafter referred to as buildings, etc.) located along the railways, etc.; places a three-dimensional construction limit model of a vehicle having a length corresponding to the length of all or part of the building, etc., whose outer surface is defined by a surface equation, on the railways, etc. data; and calculates the distance between the boundary surface of the construction limit space area determined in accordance with the three-dimensional construction limit model placed on the railways, etc. data and the building, etc., based on the three-dimensional point cloud data.

[0024] Here, "building" refers to a building or structure, such as a station, constructed for a railway project. However, the meaning is not limited by the definition in the Building Standards Act, but rather refers to buildings in general in a broad sense. "Structure" includes, for example, signaling equipment and power supply equipment, and structures defined in the Building Standards Act are included in "buildings or structures." Track data represents tracks on data composed of data other than point cloud data, and is configured as, for example, line data. Furthermore, "placement" of a vehicle-type 3D construction limit model relative to track data includes a configuration in which a three-dimensional model is created in advance and placed on the track data, and also includes a configuration in which a 2D cross-sectional model of a vehicle-type model is placed in a predetermined position on the track data and then placed on the track data using surfacing processing.

[0025] Furthermore, based on the 3D point cloud data, it is possible to generate a 3D measurement image from a viewpoint where the whole or part of a building, etc. can be recognized, and to display a visualized image in which the calculated separation distance information is visualized and overlaid on the 3D measurement image. Here, "visualization" means making the separation distance visually recognizable by using colors, shading, drawing symbols using figures, etc., rather than simply expressing the separation distance as characters (numbers, etc.).

[0026] Such a method for measuring and displaying construction gauges can be implemented by software processing using a program, and can also be configured as a system. [Effects of the Invention]

[0027] According to the present invention, in measuring the construction gauge of a building or the like, it is possible to obtain measurement results that are useful to the operator while utilizing three-dimensional point cloud data. [Brief explanation of the drawings]

[0028] [Figure 1] 1 is a schematic block diagram of a construction gauge measurement system according to a first embodiment. [Figure 2]FIG. 1 is a diagram showing a laser scanner for acquiring three-dimensional point cloud data. [Figure 3] FIG. 1 is a diagram illustrating an example of an image of a station obtained from 3D point cloud data acquired by a laser scanner. [Figure 4] FIG. 10 is a diagram illustrating an example of an image of another station obtained from 3D point cloud data acquired by a laser scanner. [Figure 5] FIG. 10 is a flow diagram of the construction gauge measurement process including the display process. [Figure 6] This is a perspective view of a vehicle-type 3D model. [Figure 7] This is a front view of a vehicle-type 3D model. [Figure 8] FIG. 10 is a diagram showing a 3D measurement image displayed during marking in a state in which a vehicle-shaped 3D model is placed on railroad data. [Figure 9] This is a diagram showing the state in which a vehicle-type 3D model is installed on railway data. [Figure 10] This figure shows a visualized image of the construction limit measurement results superimposed on a 3D measurement image. [Figure 11] FIG. 10 is a diagram schematically showing a cross section when track data RM is determined for a three-dimensional measurement image obtained from a track R in which cant occurs. [Figure 12] This is a diagram showing the position of the construction limit space area N1 taking cant into consideration. [Figure 13] This figure shows a visualized image of the results of construction gauge measurements based on an articulated vehicle model created by connecting multiple vehicle-shaped 3D models M1 without any gaps between them. [Figure 14] FIG. 10 is a diagram showing a vehicle-shaped two-dimensional cross-sectional model in the second embodiment. [Figure 15] FIG. 10 is a diagram showing a flow of model placement corresponding to step S103 in the first embodiment. [Figure 16] FIG. 10 is a diagram showing a 3D measurement image in which vehicle-shaped 2D cross-sectional models are placed at predetermined intervals on a railroad track. [Figure 17] FIG. 1 is a perspective view showing a virtual vehicle-shaped 3D model placed on a railroad track. [Figure 18] This is a plan view showing a virtual 3D vehicle model placed on a railroad track from above. [Figure 19] FIG. 10 is a diagram showing a construction gauge measurement of a construction gauge space area and a station based on the vehicle-shaped 3D model of one vehicle shown in the first embodiment. [Figure 20] This figure shows a visualized image of the construction gauge measurement based on the virtual vehicle-type 3D model shown in Figures 18 and 19. DETAILED DESCRIPTION OF THE INVENTION

[0029] Hereinafter, the construction gauge measurement system according to the present embodiment will be described with reference to the drawings. First, the construction gauge measurement system according to the first embodiment will be described with reference to Figs.

[0030] 1 is a schematic block diagram of a construction gauge measurement system according to a first embodiment. The construction gauge measurement system is connected to a database DB, a monitor MT, and an input operation unit KY, and includes a track setting unit 20, a three-dimensional (hereinafter also referred to as 3D) model placement unit 30, a construction gauge measurement unit 40, a display processing unit 50, and a point cloud image generation unit 60.

[0031] The construction gauge measurement system 10 can be configured by a computer such as a server, and the track setting unit 20, 3D model placement unit 30, construction gauge measurement unit 40, and display processing unit 50 can each be configured as a signal processing circuit. Furthermore, the construction gauge measurement system 10 can make each processing circuit function using a program stored in a memory (not shown), and can perform construction gauge measurement using software, firmware, or a combination of these.

[0032] In the database DB, data on the railway model, 3D point cloud data, and data on the construction gauge model are stored in predetermined storage areas DB1, DB2, and DB3, respectively. The construction gauge measurement system 10 acquires each piece of data as needed and temporarily stores it in a data input unit (not shown).

[0033] The track model data and the construction gauge model data are used in the track setting process and model placement process in the construction gauge measurement process (arithmetic process) described below. Meanwhile, the 3D point cloud data is stored as point cloud data acquired by 3D measurement. In this embodiment, the 3D point cloud data for the station (platform) that is the target of 3D measurement in the construction gauge measurement is acquired in advance before the construction gauge measurement process.

[0034] FIG. 2 is a diagram showing a laser scanner for acquiring three-dimensional point cloud data.

[0035] The Laser Scanner LS is configured as a 3D measurement device that acquires 3D point cloud data by emitting a laser in all directions while simultaneously taking photographs with a camera. In 3D measurement, multi-scanning is performed so that the acquired 3D point cloud data can be used to generate and display a 3D image that captures the entire station as if it were a bird's-eye view of the station being measured.

[0036] 3 and 4 are diagrams illustrating images of a station obtained from three-dimensional point cloud data acquired by a laser scanner LS.

[0037] Three-dimensional point cloud data is a collection of point data, each having three-dimensional coordinate values, allowing the distance between any two points to be calculated. Furthermore, coordinate transformations (coordinate system transformations) such as arbitrary rotation and translation can be performed on point cloud data in three-dimensional space. Therefore, regardless of the location of the laser scanner LS, it is possible to generate and display a three-dimensional image from any viewpoint. The point cloud image generator 60 generates a three-dimensional image (hereinafter referred to as a three-dimensional measurement image) that can be viewed stereoscopically from a viewpoint specified by the operator or a predetermined viewpoint, as an image based on the point cloud.

[0038] By performing 3D measurements with the laser scanner LS at the installation location shown in Figure 2, 3D point cloud data can be obtained by scanning and photographing the support pillars C that support the platform shed (not shown in Figure 2) and the tracks R. As a result, 3D images of the measurement results, such as those seen from one end of the station and the other end of the station, can be generated and displayed, as shown in Figures 3 and 4.

[0039] For a 3D measurement image, a 3D coordinate system is defined with a predetermined position as its origin. For example, the X and Y axes are defined as directions perpendicular to each other along a horizontal plane, and the Z axis is defined as a direction perpendicular to the X and Y plane. Each point cloud data has coordinate values ​​based on the same units (e.g., mm) as those used in real 3D space. Furthermore, coordinate system transformation can be used to convert the coordinate values ​​to those with a desired position as the origin.

[0040] The 3D point cloud data acquired by the laser scanner LS is transmitted to a terminal (not shown) or the like, and stored in a database DB via the terminal or the like. The 3D point cloud data stored in the database DB is subjected to registration processing (positioning) by the laser scanner LS or the terminal or the like. In addition to the 3D measurement, measurement work is performed in advance using a 3D measuring device such as a total station to acquire information about the vertical direction and / or horizontal plane, and the information is stored in the database DB.

[0041] After the 3D measurement, the operator operates the input operation unit KY while looking at the monitor MT to perform the construction gauge measurement work. In response to the operator's input operations, the construction gauge measurement system 10 sets the track for model placement, places the rolling stock type construction gauge model on the track, measures the construction gauge, and displays a visualized image that visualizes the distance between the construction gauge and the station. This will be described in detail below.

[0042] FIG. 5 is a flow diagram of the construction gauge measurement process including the display process.

[0043] When an operator operates an input operation unit KY such as a keyboard, the construction gauge measurement system 10 acquires three-dimensional point cloud data from the database DB (S101). Then, based on the three-dimensional point cloud data, a railroad track (hereinafter referred to as railroad track data) on the data is set (S102).

[0044] The track R is made up of straight and curved sections, and the rail width and cross-sectional shape are determined according to the straight and curved sections. Therefore, it is possible to determine track data based on point cloud information corresponding to the track R contained in the 3D point cloud data. The track model stored in the database DB is a data model constructed based on track information corresponding to each of the straight and curved sections, and is represented here as line data having a predetermined width.

[0045] The track setting unit 20 extracts track information (point cloud data) contained in the 3D point cloud data by matching it with a track model stored in the database DB, and determines the track data. Various calculation methods can be applied to set the track data. Furthermore, a neural network (NN) such as deep learning may be used to automatically recognize tracks in a 3D image.

[0046] After the track data is set, a vehicle-type 3D model as a graphic model is placed on the track data (S103) (Note that, hereinafter, the track data may be simply referred to as "track"). The placement of the vehicle-type 3D model on the track data will be described below with reference to Figures 6 to 9.

[0047] Fig. 6 is a perspective view of the vehicle-shaped 3D model, and Fig. 7 is a front view of the vehicle-shaped 3D model.

[0048] The vehicle-type 3D model is a graphic model that serves as the basis for determining the construction limit area, and is obtained from the database DB. The vehicle-type 3D model M1 has an exterior shape that allows for some margin in the vehicle's exterior frame, and is configured as a linear, three-dimensional model. The vehicle-type 3D model is also configured as a model with a length shorter than the entire station PF, and in this case is configured as a three-dimensional graphic model corresponding to the length of one vehicle.

[0049] The vehicle-shaped 3D model M1 has surface data, which is information about its outer surface, and its outer surface can be defined by an equation that geometrically represents the surface in a defined three-dimensional coordinate system. The vehicle-shaped 3D model M1 has, as surface data, information about multiple flat or curved surfaces that are defined in accordance with the edges of the vehicle-shaped 3D model (end points of the cross-sectional shape).

[0050] Furthermore, because the construction gauge area of ​​the vehicle-type 3D model M1 is defined according to the vehicle type, its outer surface is composed of multiple surfaces. In Figures 6 and 7, parts of the outer surface, each with different surface data (equations), are indicated by the symbols S1 and S2. Furthermore, four wheel data WS1 to WS4 are defined for the vehicle-type 3D model M1 according to the rail width of the track R.

[0051] FIG. 8 is a diagram showing a three-dimensional measurement image displayed during marking in a state in which the vehicle-type 3D model M1 is placed on the track data.

[0052] The point cloud image generation unit 60 generates a 3D measurement image viewed from above the station PF in response to input operations by the operator. The display processing unit 50 displays the 3D measurement image as a point cloud image on the monitor MT, as shown in Fig. 8. The station PF represented by the 3D point cloud data is displayed as a building constructed along a straight section of the track R, and the track data RM is also defined as straight line data. Note that the same symbols are used for the actual station, roof, and pillars shown in Fig. 2 and the station, roof, and pillars displayed in the 3D measurement image.

[0053] The operator aligns the positions of the wheel data WS1 to WS4 of the vehicle-type 3D model M1 with the positions of the track data RM, i.e., the positions of the inner rail data RM1 and outer rail data RM2, and places the vehicle-type 3D model M1 on the track. Specifically, the operator marks the track RM and moves the vehicle-type 3D model. The display processing unit 50 executes display processing in response to the input operation, displays the markings, changes the display location of the vehicle-type 3D model, etc.

[0054] As shown in FIG. 8, four markers M1 to M4 are displayed superimposed on the track data RM in accordance with the positions of the wheel data W1 to W4. The markers M1 to M4, which are drawn by the operator's marking input operation, serve as indicators to the mounting location of the vehicle-type 3D model M1. Here, the vehicle-type 3D model M1 is displayed semi-transparently, and the wheel data W1 to W4 are displayed in color so that the wheel data W1 to W4 can be identified. However, in FIG. 8, the vehicle-type 3D model M1 itself is not shown.

[0055] FIG. 9 is a diagram showing a state in which the vehicle-type three-dimensional model M1 is loaded onto the railroad track data RM.

[0056] The operator performs an operation (such as a mouse operation involving clicking) to align the positions (center lines) of the wheels W1 to W4 of the vehicle-type 3D model M1 with the positions of the markers M1 to M4 on the track RM (the center lines of the track data RM), thereby mounting the vehicle-type 3D model M1 on the track RM. At this time, the vehicle-type 3D model M1 may be positioned while overlooking the entire or part of the station PF and displaying a 3D measurement image from a perspective that views the vehicle-type 3D model M obliquely on the monitor MT.

[0057] After the 3D vehicle-type model M1 is placed on the track RM, a construction gauge measurement is performed (S104). Specifically, the separation distance between the construction gauge area determined based on the vehicle-type 3D model M1 and the station PF is measured for each of the 3D point cloud data of the station PF.

[0058] As described above, the track data RM is expressed as linear line data, and the vehicle-type 3D model M1 is configured to have a vehicle-type external shape. Therefore, assuming that there is no cant with respect to the track R (no inclination occurs between the inner rail MR1 and the outer rail MR2), the vehicle-type 3D model placed on the track data RM can be defined as the construction gauge region N1 as is, as shown in Figure 9. Hereinafter, to clarify that the construction gauge region is a spatial region extending along the track RM, it will be referred to as the "construction gauge spatial region."

[0059] Because the vehicle-type 3D model M1 has surface data, the boundary surface (outer surface) of the construction limit spatial region (= vehicle-type 3D model) N1 placed on the track RM can be expressed by the equations of the 3D coordinate system defined in the 3D measurement image. Therefore, the distance from each point of the station (PF) represented by the 3D point cloud data to the closest point (also called the closest point) on the boundary surface of the construction limit spatial region N1 can be calculated as the separation distance.

[0060] Once the separation distance calculation process (construction limit measurement) is performed, a process is executed to display a visualized image of the measurement results (S105). Here, the 3D measurement image is colored in a way that changes color depending on the degree of separation distance. For example, parts that interfere with the construction limit and / or building elements that are close to the construction limit spatial region N1 can be displayed in red or yellow, while areas with sufficient separation distance can be displayed in blue or no color. Note that, because the boundary surface of the construction limit spatial region N1 is expressed by a surface equation, it is possible to distinguish point cloud data that exists inside the construction limit spatial region N1 from the measurement results, and these can be identified as parts that interfere with the construction limit.

[0061] It is also possible to visualize the degree of construction gauge deviation and obstacles by using other methods, such as filling in the image or drawing symbols such as circles or polygonal marks. Furthermore, it is also possible to visualize the distance deviation by displaying it in a frame. It is also possible to display the numerical values ​​and other characters as they are, for example, in a table, without using an image.

[0062] Figure 10 is a diagram showing a visualized image of the separation distance, i.e., the results of the construction limit measurement, superimposed on a 3D measurement image. The 3D measurement image is an image viewed from an oblique viewpoint that allows a bird's-eye view of the entire construction limit spatial area N1 located at a specified location in the station PF, and allows an overall recognition of the degree of separation from the station PF in relation to the entire construction limit spatial area N1.

[0063] Instead of measuring the entire structure gauge spatial region N1, only a portion designated by an operator or the like may be the target of structure gauge measurement, and the distance between that portion and the building elements of the station PF that face or are close to it may be measured and calculated by measuring the structure gauge. For example, the edge of the roof (platform shed) PM of the station PF may be designated, and the distance between that edge and the building elements may be visualized.

[0064] The above-mentioned construction gauge measurement does not take cant into account, as it targets station PFs along straight sections of track R. However, there are cases where station PFs are installed in track sections that include curved sections with slight curvature. In such cases, cant can be obtained from 3D measurement images, and a construction gauge spatial region that takes cant into account can be defined.

[0065] FIG. 11 is a diagram schematically showing a cross section when track data RM is determined for a three-dimensional measurement image obtained from a track R where a cant occurs.

[0066] As mentioned above, in addition to the 3D measurements using the laser scanner LS, the vertical direction is also measured using a measuring instrument. Therefore, each point in the acquired 3D point cloud data has 3D coordinate values ​​that are aligned with the horizontal and vertical directions. If track data RM consisting of the inner rail RM1 and outer rail RM2 is set as line data, the difference in elevation between the inner rail RM1 and outer rail RM2 can be calculated from the coordinate values ​​of the point cloud data on that line data (for example, the position of the center of the track width), and the cant can then be determined.

[0067] Even if measurements in the vertical direction are not taken, cant can be calculated by assuming that the plane perpendicular to the support pillar C of the station PF is the horizontal plane. The plane perpendicular to the support pillar C of the station PF can be calculated from a 3D measurement image. Cant may also be calculated using a calculation method such as those shown in any of Prior Art Documents 1 to 7.

[0068] Figure 12 shows the position of the structure limit space area N1 taking cant into account. The structure limit space area is rotated by an angle α with respect to the horizontal plane, and the structure limit measurement is performed at that position. Here, the axis of rotation is based on the center line of the bottom surface of the structure limit space area N1. However, for convenience, in Figure 12, the structure limit space area N1 is represented as a two-dimensional cross-sectional model.

[0069] In the first embodiment, a vehicle-type 3D model M1 having the length of one vehicle is used, but the station PF is constructed to match the overall length of a train consisting of multiple coupled vehicles. Therefore, a 3D model in which multiple vehicle-type 3D models M1 are connected may be used to measure the construction gauge and display a visualized image.

[0070] Figure 13 shows a visualized image of the results of construction gauge measurements based on an articulated vehicle model created by connecting multiple vehicle-type 3D models M1 without any gaps between them. By defining the construction gauge spatial region N1 based on this articulated vehicle model, it is possible to visually recognize the separation distances and construction gauge obstacles throughout the entire station PF.

[0071] As described above, according to the first embodiment, 3D measurement is performed using a laser scanner LS from above the station PF. Then, track data RM is set based on a 3D measurement image obtained from the 3D point cloud data, and a vehicle-type 3D model M1 created and prepared in the database DB is placed on the track data RM.

[0072] The system calculates the distance between the construction limit space area N1, which is defined according to the vehicle-type 3D model M1 placed on the track RM, and the station PF, and displays a visualized image in which the distance information is visualized and overlaid on a 3D measurement image from a viewpoint that allows a bird's-eye view of the entire construction limit space area N1 (vehicle-type 3D model M1).

[0073] In this embodiment, unlike conventional methods, point cloud data existing within the construction limit area frame, which is a two-dimensional cross-sectional model, is extracted from the viewpoint in the direction of travel of the traveling vehicle, and obstacle detection is not performed. Instead, a visualized image is displayed that allows the distance between the outer surface of the construction limit space area N1 and the station PM to be recognized, based on a three-dimensional measurement image that provides a bird's-eye view and oblique view of the entire station PM that is the target of construction limit measurement.

[0074] This makes it possible to grasp not only the distance and obstruction between a specific point in station PM and the vehicle, but also the state of distance between the outer surface of the vehicle and the entire platform shed extending along track R. In other words, it becomes possible to recognize the degree of distance between the construction gauge area side and the measurement target side on a "surface-by-surface" basis.

[0075] On the other hand, in the construction gauge measurement according to this embodiment, a vehicle-type 3D model M1 with surface data is created and placed on the track data RM to define the construction gauge spatial region N1. The separation distance from the station PF represented by the three-dimensional point cloud data is calculated as the distance between the surface of the construction gauge spatial region N1 (vehicle-type 3D model M1) and each point of the point cloud data, so the separation distance can be measured with high accuracy.

[0076] Furthermore, since the vehicle-type 3D model M1 can be placed on the track at the operator's desired location, the vehicle-type 3D model M1 can be placed at the location where the operator wants to recognize the degree of separation, and it is also possible to perform construction gauge measurements by moving the location of the vehicle-type 3D model M1 intermittently or by small distances.

[0077] In this embodiment, instead of mounting a 3D measurement device on a running vehicle and acquiring 3D point cloud data as in the past, 3D measurements are performed at a location outside the vehicle (here, on the station platform). Even with this type of 3D measurement, it is possible to recognize gaps and obstacles for the entire station PM that is the target of construction gauge measurement.

[0078] Furthermore, the position of railway tracks can change due to movement or subsidence of the trackbed caused by construction work, etc. In contrast, stations are durable, stable structures that are designed to be used for many years without being demolished, so by periodically conducting 3D measurements from the station, it is possible to measure the construction limit taking into account changes in the position of the tracks and identify gaps and obstacles.

[0079] Furthermore, because no moving vehicles are used, there is no need to correct the 3D point cloud data to take into account the swaying of the moving vehicle, and 3D measurements can be performed on the platform during times of good weather, eliminating the need to perform 3D measurements at night.Furthermore, there is no need to place a 3D measuring device on the rails, so track closure procedures are unnecessary.

[0080] In addition, 3D measurements may be performed at locations other than the platform, as long as 3D point cloud data can be obtained from a viewpoint that allows for the recognition of distances and obstacles relative to the entire station PM that is the target of the construction gauge measurement.

[0081] In this embodiment, the vehicle-type 3D model M1 is placed on the track RM through input operations by an operator, but model placement may also be performed automatically. For example, the placement location of the vehicle-type 3D model M1 may be determined in advance with respect to the track model stored in the database DB. Alternatively, point cloud data of stations PF may be automatically extracted from the 3D measurement image, and the placement location of the vehicle-type 3D model M1 with respect to the set track data RM may be determined in accordance with the position of the station PF.

[0082] Alternatively, the vehicle-type 3D model M1 may be placed in the track data RM by semi-automated processing. For example, an operator may perform an input operation such as tracing the track on a 3D measurement image using a mouse or pen, thereby automatically extracting point cloud information representing the track R and setting the track data RM. Then, the vehicle-type 3D model M1 may be placed in the track data RM.

[0083] Next, a construction gauge measurement system according to a second embodiment will be described with reference to Figures 14 to 20. Unlike the first embodiment, the second embodiment uses a two-dimensional construction gauge model as a base and generates a virtual vehicle-shaped 3D model having an external shape different from that of an actual vehicle.

[0084] FIG. 14 is a diagram showing a vehicle-type 2D cross-sectional model in the second embodiment. The vehicle-type 2D cross-sectional model m2 is configured as a 2D model having a contour corresponding to a conventional construction gauge measurement area. The contour shape here is the same as the 2D cross-sectional shape of the vehicle-type 3D model in the first embodiment. The database DB stores the vehicle-type 2D cross-sectional model m2 as data of the construction gauge model.

[0085] In the second embodiment, station PF is constructed along a railway track R whose trajectory includes curved sections, and roof PM is also formed along the path of the railway track R. In order to generate a vehicle-type 3D model for such a railway track R that includes curved sections, multiple vehicle-type 2D cross-sectional models m2 are placed on the railway track R, and a continuous outer surface is formed to connect these, thereby generating a virtual vehicle-type 3D model. This will be described in detail below.

[0086] Fig. 15 is a diagram showing a flow of model placement corresponding to step S103 in the first embodiment. Fig. 16 is a diagram showing a 3D measurement image in which vehicle-type 2D cross-sectional models m2 are placed at predetermined intervals on a railway RM. Note that here, a point cloud image is shown schematically.

[0087] After acquiring the vehicle-type 2D cross-sectional model m2 from the database DB (S201), the vehicle-type 2D cross-sectional model m2 is modified to match the placement location (S202). In the second embodiment, multiple vehicle-type 2D cross-sectional models m2 are placed at predetermined intervals with respect to the track data MR. Here, the vehicle-type 2D cross-sectional models m2 are placed by an operator's input operation or automatically to match the positions of the support pillars C installed throughout the station PF.

[0088] Since the track R has a curved section along station PF, the track image RM also has a curved section. Therefore, as explained in the first embodiment, cant (the inclination between the inner rail RM1 and the outer rail RM2) occurs. In addition, a transition curve is provided in the curved section, and a complex track trajectory (route) is adopted in which the curve radius gradually decreases. Furthermore, the curved section of the track causes deviation in the running vehicle. To address this, it is necessary to expand the construction gauge area.

[0089] In the second embodiment, the contour shape of the vehicle-shaped 2D cross-sectional model m2 is changed and enlarged in consideration of cant, transition curves, and deviation. Cant can be calculated in the same manner as in the first embodiment. Then, the vehicle-shaped 2D cross-sectional model m2 is rotated based on the calculated cant.

[0090] Furthermore, for the vehicle-shaped 2D cross-sectional model m2 that takes transition curves and deviations into account, information such as the curve radius (radius of the arc portion) can be stored in advance in a database or the like, and the information can be read out to change or expand the contour shape of the vehicle-shaped 2D cross-sectional model m2. For example, the height and width directions shown in FIG. 7 can be extended or scaled (see, for example, FIG. 6 of JP 2005-271717 A). Note that the calculation methods described in any of the prior art documents 1 to 7 may be used to change or expand the area of ​​the vehicle-shaped 2D cross-sectional model m2 based on cant, transition curves, or vehicle deviation. Furthermore, the contour shape of the vehicle-shaped 2D cross-sectional model m2 may be changed by taking slack into account using the calculation methods described in any of the patent documents 1 to 7.

[0091] After arranging multiple vehicle-shaped 2D cross-sectional models m2 at predetermined locations (S203), a continuous outer surface is formed by graphics processing to connect the vehicle-shaped 2D cross-sectional models m2 whose contour shapes have been changed, and a virtual vehicle-shaped 3D model is generated (S204).

[0092] The graphics processing should be a method that can represent the outer surface (including approximation) using a surface equation in three-dimensional coordinate space and that can form a continuous surface shape. In particular, it should be a configuration that performs surface processing that can form an outer surface whose shape changes smoothly. Here, surface processing using a blend (loft) feature is used.

[0093] Fig. 17 is a perspective view showing a virtual vehicle-type 3D model M2 placed on a railway track RM. Fig. 18 is a plan view showing the virtual vehicle-type 3D model M2 placed on a railway track RM from above. Both are displayed in a 3D measurement image, which is a point cloud image. Note that the symbol M2 is not shown in Figs. 17 and 18.

[0094] As shown in Figure 18, station PF is installed along the complex trajectory of the tracks. Therefore, the contour shapes of the vehicle-type 2D cross-sectional models m2, which are arranged to match the positions of each support pillar, do not match. A continuous outer surface shape connecting multiple vehicle-type 2D cross-sectional models m2 is formed by surfacing processing, and a virtual vehicle-type 3D model M2 is generated. The virtual vehicle-type 3D model M2 is configured as a 3D model whose length extends along the crossover track RM for almost the entire station PF.

[0095] The virtual vehicle-type 3D model M2 generated in this way is a 3D model generated while already installed on the track RM, and is therefore defined as the construction gauge space area as is. Then, as in the first embodiment, the construction gauge measurement process is executed and a visualized image is displayed.

[0096] Unlike an actual train, the virtual vehicle-type 3D model M2 does not have a constant two-dimensional cross-sectional shape, but has an outer surface shape in which the contour expands and contracts in a complex manner along the train direction.On the other hand, the virtual vehicle-type 3D model M2 is a 3D model based on the vehicle-type two-dimensional cross-sectional model m2, whose contour shape has been modified to take into account cant, transition curves, and vehicle deviation, and is therefore configured as a 3D model that generates construction gauge measurements that are stricter than the actual distance.

[0097] FIG. 19 is a diagram showing the construction gauge measurement of a construction gauge space area and a station based on the vehicle-type 3D model of one vehicle shown in the first embodiment.

[0098] In Figure 19, station PF is installed along track RM, which is a curved section with a constant curvature. In this case, the separation distance varies depending on the position of the construction limit space region N1 (vehicle-type 3D model) (see symbols A and B).

[0099] However, the virtual vehicle-type 3D model M2 generated in the second embodiment has a curved (snake-like) outer surface that matches the curvature of the track R, regardless of the trajectory of the track R, thereby suppressing the occurrence of a difference in distance due to the model arrangement. Furthermore, the virtual vehicle-type 3D model M2 has a length that spans almost the entire station PF, so the distance from the vehicle can be captured in a visualized image over the entire roof PM of the station PF.

[0100] Figure 20 shows a visualized image of the construction gauge measurement based on the virtual vehicle-type 3D model M2 shown in Figures 18 and 19. Appropriate clearance measurements and obstruction detection can be performed even for station PF, which is constructed along track R, including curved sections. In particular, the virtual vehicle-type 3D model M2 is configured as a 3D model in which clearances are stricter than in reality (closer values) depending on the measurement location, allowing for reliable safety measures to be taken.

[0101] In the second embodiment, the vehicle-shaped 2D cross-sectional model m2 is positioned to match the position of the support pillar C of the station PF. However, considering that the coordinate values ​​of the 3D point cloud data are obtained after measuring the vertical direction with a measuring device as described above, it is preferable to position the vehicle-shaped 2D cross-sectional model m2 (separately) to match the positions of the straight and curved sections (particularly arc sections) of the track RM. In this case, the cant that matches the placement location, the curve radius of the curved section (arc section), and the like can be stored in the database DB. This prevents the degree of area change of the vehicle-shaped 2D cross-sectional model m2 based on cant, etc., from changing suddenly before and after the change, allowing the outer surface of the virtual vehicle-shaped 3D model to have a smoother curve.

[0102] In the first and second embodiments, a virtual 3D vehicle model based on a vehicle-type 3D model or a 2D vehicle model is placed on the track for a 3D measurement image, which is a point cloud image, and a visualized image based on the 3D measurement image is also displayed. However, the 3D point cloud data may be converted into voxel data, mesh data, etc., and a 3D image may be generated and displayed using data different from the point cloud data. On the other hand, for construction gauge measurement, the separation may be calculated based on the 3D point cloud data. This makes it possible to determine the separation with high accuracy and provide a 3D image that is easy for the operator to view.

[0103] In the first and second embodiments, the lower construction gauge measurement is performed for stations installed along the tracks, but it can also be applied to railway-related facilities such as overhead lines and structures such as signal equipment.

[0104] Furthermore, the above-mentioned construction gauge measurement can be applied not only to vehicles traveling on railroad tracks but also to vehicles traveling on roads. In this case, the track center line is identified from the 3D measurement image, and a three-dimensional vehicle-shaped 3D model with widths to both edges of the road cross section and a height according to the vehicle height (for example, the maximum limit of 3.8 m) is created as a 3D construction gauge model.

[0105] This allows the distance from the median strip to be measured, and the distance from the construction limit space area to be visually recognized on a "surface-by-surface" basis. Construction limit measurements are particularly effective on expressways and other roads that take into account Level 4 autonomous driving. Also, unlike railway vehicles, there are various vehicle configurations, so it is possible to prepare multiple three-dimensional vehicle-shaped 3D models and select one to measure the construction limit. If the driving surface is inclined, the 3D construction limit model can be modified to match the inclined surface, using methods such as those shown in Patent Publication No. 7162779.

[0106] Furthermore, the technology can be applied not only to vehicles that run on roads, but also to a variety of other vehicles, such as streetcars that run in urban areas and DMVs (Dual Mode Vehicles) that run on both rails and roads. [Explanation of symbols]

[0107] 10 Construction Gauge Measurement System 20 Track setting section 30 3D model placement section 40 Construction Gauge Measurement Section 50 Display processing section 60 Point cloud image generation unit C-post DB Database LS laser scanner M1 vehicle-type 3D model (3D construction limit model) M2 Virtual vehicle-shaped 3D model (3D construction limit model) m2 Vehicle-shaped 2D cross-sectional model N1 Construction limit space area PF Station PM roof (platform roof) R track RM track data

Claims

1. determining data on railways or roads (hereinafter referred to as railways, etc.) and buildings or structures (hereinafter referred to as buildings, etc.) provided along the railways, etc. from three-dimensional point cloud data obtained by three-dimensionally measuring the railways, etc.; a three-dimensional structure gauge model of a vehicle having a length corresponding to the length of the whole or part of the building, etc., is placed on the data of the railway, etc., having a trajectory including a curved portion; A construction gauge measurement method for calculating a separation distance between a boundary surface of a construction gauge space area determined in accordance with a three-dimensional construction gauge model arranged in data of the railway line, etc., and the building, etc., based on three-dimensional point cloud data, A method for measuring a construction gauge, characterized in that a three-dimensional construction gauge model having a curved outer surface formed according to the curvature of the data of the railway line, etc., is placed on the data of the railway line, etc.

2. determining data on railways or roads (hereinafter referred to as railways, etc.) and buildings or structures (hereinafter referred to as buildings, etc.) provided along the railways, etc. from three-dimensional point cloud data obtained by three-dimensionally measuring the railways, etc.; a three-dimensional construction gauge model of a vehicle having a length corresponding to the length of the whole or part of the building, etc., the three-dimensional construction gauge model having an outer surface defined by a surface equation, is placed on data of a railway, etc.; The distance between the boundary surface of the construction limit space area, which is determined in accordance with the three-dimensional construction limit model arranged in the data of the railway line, etc., and the building, etc., is calculated based on the three-dimensional point cloud data. A method for measuring construction gauges.

3. The method for measuring construction limits described in claim 1, characterized in that the three-dimensional construction limit model is generated by connecting multiple two-dimensional cross-sectional models placed on the data of the railway line, etc. with a continuous outer surface by surface processing using a loft feature.

4. The method for measuring construction limits described in claim 3, characterized in that the plurality of two-dimensional cross-sectional models are placed on the data of the railway, etc., after at least one of enlarging and rotating the plurality of two-dimensional cross-sectional models based on at least one of the curve radius and cant of the data of the railway, etc.

5. The method for measuring construction limits according to claim 2, characterized in that the three-dimensional construction limit model has, as surface data, a plurality of plane information or curved surface information defined by the equation of the surface in accordance with its edges.

6. generating a three-dimensional measurement image from a viewpoint that allows the entire or part of the building, etc. to be recognized based on the three-dimensional point cloud data; A method for measuring construction limits according to any one of claims 1 to 5, characterized in that a visualized image is displayed in which the calculated separation distance information is visualized and superimposed on the three-dimensional measurement image.

7. In a computer, determining data of the railway or road (hereinafter referred to as the railway, etc.) from three-dimensional point cloud data obtained by three-dimensionally measuring the railway or road and buildings or structures (hereinafter referred to as the buildings, etc.) provided along the railway, etc.; a step of arranging a three-dimensional construction gauge model of a vehicle having a length corresponding to the length of the whole or part of the building onto data of the railway line or the like having a trajectory including a curved portion, the three-dimensional construction gauge model having a curved surface formed on its outer surface according to the curvature of the data of the railway line or the like onto the data of the railway line or the like; a step of calculating a distance between the boundary surface of a construction limit space area determined in accordance with a three-dimensional construction limit model arranged in the data of the railway line, etc., and the building, etc., based on the three-dimensional point cloud data; A program characterized by executing the following.

8. In a computer, determining data of the railway or road (hereinafter referred to as the railway, etc.) from three-dimensional point cloud data obtained by three-dimensionally measuring the railway or road and buildings or structures (hereinafter referred to as the buildings, etc.) provided along the railway, etc.; a step of arranging a three-dimensional construction gauge model of a vehicle having a length corresponding to the length of the entire or part of the building, the three-dimensional construction gauge model having an outer surface defined by a surface equation, onto data of a railway line or the like; a step of calculating a distance between the boundary surface of a construction gauge space area defined in accordance with a three-dimensional construction gauge model of a vehicle arranged in the data of the track, etc., and the building, etc., based on the three-dimensional point cloud data; A program characterized by executing the following.

Images