Cutting face analysis system and cutting face analysis method

Abstract

To provide a cutting face analysis system 1 and a cutting face analysis method capable of easily evaluating a cutting face F using a three-dimensional model 37 of the cutting face F.SOLUTION: A cutting face analysis system 1 comprises: point cloud data acquisition means for acquiring point cloud data of a cutting face F; three-dimensional model generation means for generating a three-dimensional model 37 of the cutting face F based on the point cloud data; first normal vector calculation means for calculating a first normal vector V passing through a center of gravity G of a mesh plane 37a; partition data generation means for generating partition data 39A on the three-dimensional model 37; second normal vector acquisition means for acquiring, for each partition 39s, a normal vector calculated based on the first normal vector V included in the partition 39s of the partition data 39A as a second normal vector within the partition 39s; and output means for outputting colored partition data 39B in which the partitions 39s are colored with a predetermined color corresponding to the second normal vector.SELECTED DRAWING: Figure 7

Description

Technical Field

[0001] The present invention relates to a face analysis system and a face analysis method for analyzing, for example, a tunnel face and assisting in the evaluation of the face.

Background Art

[0002] At a tunnel excavation site, it is necessary to observe the state of the face daily as excavation progresses and to grasp the locations where peeling is predicted and the geological state in the excavation direction. At this time, it is desirable for an observer to approach the face and visually observe the state of the face and measure the strike and dip of the discontinuity plane using a clinometer, but there is a problem that it is difficult to ensure the safety of the observer.

[0003] Therefore, various techniques have been proposed that enable the state of the face to be observed without the observer approaching the face. For example, Patent Document 1 discloses an apparatus that extracts a discontinuity plane that appears on a tunnel face based on a three-dimensional model obtained by measuring the tunnel face with a three-dimensional laser scanner, and supports the evaluation of the face by the user by color-coding and displaying the mesh plane of the three-dimensional model corresponding to the discontinuity plane.

[0004] By the way, since the three-dimensional model of the face can reproduce even the fine irregularities of the face in the virtual space, there is an advantage that the state of the face can be accurately confirmed. However, when a user views a three-dimensional model with color-coded mesh planes as in Patent Document 1 in a front view, there is a possibility that the user may not be able to visually recognize the color-coded mesh planes depending on the orientation of the mesh planes.

[0005] For this reason, the user needs to evaluate the state of the face while checking the color-coded three-dimensional model from multiple directions. Then, in Patent Document 1, there is a possibility that it may be time-consuming for the user to evaluate the state of the face, and there is room for improvement.

Prior Art Documents

Patent Documents

[0006] [Patent Document 1] Japanese Patent No. 5986362 [Summary of the Invention] [Problems to be Solved by the Invention]

[0007] In view of the above problems, an object of the present invention is to provide a face analysis system and a face analysis method that facilitate the evaluation of a face using a three-dimensional model showing the shape of the face. [Means for Solving the Problems]

[0008] This invention is a face analysis system for analyzing the face of a tunnel and assisting in the evaluation of the face, comprising: point cloud data acquisition means for acquiring point cloud data showing the shape of the face; three-dimensional model generation means for generating a three-dimensional model composed of triangular mesh planes having adjacent points of the point cloud data as vertices; first normal vector calculation means for calculating a first normal vector along the normal direction passing through the centroid of the mesh plane; section data generation means for generating section data in which sections of a predetermined size are two-dimensionally arranged on the three-dimensional model; second normal vector acquisition means for acquiring, for each section, one normal vector obtained based on the first normal vector included in the section of the section data as the second normal vector in the section; and output means for outputting the section data in which the sections are color-coded with a predetermined color corresponding to the direction indicated by the second normal vector.

[0009] The present invention also relates to a face analysis method for analyzing the face of a tunnel and assisting in the evaluation of the face, comprising: a point cloud data acquisition step in which point cloud data acquisition means acquires point cloud data indicating the shape of the face; a three-dimensional model generation step in which three-dimensional model generation means generates a three-dimensional model composed of triangular mesh planes having adjacent points of the point cloud data as vertices; a first normal vector calculation step in which first normal vector calculation means calculates a first normal vector along the normal direction passing through the centroid of the mesh plane; a section data generation step in which section data generation means generates section data in which sections of a predetermined size are two-dimensionally arranged on the three-dimensional model; a second normal vector acquisition step in which second normal vector acquisition means acquires, for each section, one normal vector obtained based on the first normal vectors included in the section as the second normal vector in the section; and an output step in which output means outputs the section data in which the sections are color-coded with a predetermined color corresponding to the direction indicated by the second normal vector.

[0010] The above-mentioned point cloud data refers to point cloud data generated based on a moving image of the face taken, or point cloud data based on measurement data obtained by measuring the face with a three-dimensional scanner. The above-mentioned point cloud data acquisition means refers to means for acquiring point cloud data generated by point cloud data generation means, means for acquiring point cloud data based on measurement data, means for acquiring point cloud data via a communication line, or means for acquiring point cloud data via a portable storage medium.

[0011] The above-mentioned section data is data of a two-dimensional plane divided into sections such as quadrilateral or hexagonal sections, data generated separately from the three-dimensional model and overlaid on the three-dimensional model, or data generated by dividing the three-dimensional model into quadrilateral or hexagonal sections. The above-mentioned second normal vector means refers to means that does not obtain a second normal vector when there is no first normal vector included in the section.

[0012] The above face analysis system refers to a system in which the point cloud data acquisition means, the three-dimensional model generation means, the first normal vector calculation means, the section data generation means, the second normal vector acquisition means, and the output means are provided in one terminal or server. Alternatively, it refers to a system in which one or more of the point cloud data acquisition means, the three-dimensional model generation means, the first normal vector calculation means, the section data generation means, the second normal vector acquisition means, and the output means are provided in the server, and the above means other than those provided in the server are provided in the client terminal.

[0013] According to the present invention, the position and orientation of the artificial excavation surface in the three-dimensional model and the position and orientation of the surfaces other than the artificial excavation surface (for example, the discontinuous surface that appears on the face) can be replaced on the two-dimensional plane.

[0014] Therefore, for example, when the mesh plane of the three-dimensional model is color-coded, it is necessary to observe the face state by observing the three-dimensional model from multiple directions, whereas the face analysis system can enable the evaluation of the face without observing from multiple directions.

[0015] Furthermore, by the cooperation of the section data generation means and the second normal vector acquisition means, the face analysis system can obtain the second normal vector by regarding it as the normal vector of one mesh plane within the section. Thereby, since the face analysis system can output the color-coded section data while suppressing the influence of fine unevenness, the state of the face can be clarified.

[0016] Therefore, the face analysis system and the face analysis method using the same can facilitate the evaluation of the face using the three-dimensional model as compared with the case of evaluating the face using the three-dimensional model with the mesh plane color-coded.

[0017] As an aspect of the present invention, when there are a plurality of the first normal vectors within the section of the section data, the second normal vector acquisition means may be configured to acquire an average value of the plurality of first normal vectors as the second normal vector.

[0018] The second normal vector means refers to means for acquiring the first normal vector within the section as the second normal vector when there is one first normal vector within the section. According to this configuration, since it is possible to output section data that has been color-coded while reliably suppressing the influence of fine unevenness, the state of the face can be made clearer.

[0019] As another aspect of the present invention, the point cloud data acquisition means may be configured to acquire the point cloud data from video data obtained by photographing the face. According to this configuration, for example, point cloud data can be acquired from video data obtained by photographing the face with a mobile terminal equipped with a camera or a digital camera.

[0020] Thereby, the face analysis system can acquire point cloud data without installing a three-dimensional laser scanner or the like inside the tunnel, and thus the burden on the workers at the excavation site can be reduced.

[0021] As another aspect of the present invention, it may be provided with first normal vector extraction means for extracting the first normal vectors with high appearance frequency, and first normal vector deletion means for deleting the first normal vectors extracted by the first normal vector extraction means.

[0022] The first normal vectors with high appearance frequency refer to the first normal vector with the highest appearance frequency, or the first normal vectors with the appearance frequency up to the number counted from the highest appearance frequency. Or it refers to the first normal vector with the highest appearance frequency and the first normal vectors within a predetermined angular range obtained by adding a predetermined angle to the direction indicated by the first normal vector.

[0023] According to this configuration, by deleting the first normal vector with a high appearance frequency, it is possible to exclude the replacement of the artificial excavation surface with a high appearance frequency onto the two-dimensional plane. In other words, the face analysis system can replace only the position and orientation of the faces other than the artificial excavation surface onto the two-dimensional plane.

[0024] Thereby, the face analysis system can not only efficiently color-code the sections of the section data, but also exclude the artificial excavation surface from the evaluation target, so that the evaluation of the face can be made easier.

[0025] Also, as an aspect of the present invention, there may be provided first normal vector extraction means for extracting the first normal vector whose direction indicated by the first normal vector is within a predetermined angle range with respect to the tunnel axis direction, and first normal vector deletion means for deleting the first normal vector extracted by the first normal vector extraction means. The above-mentioned predetermined angle range is an angle range substantially orthogonal to the tunnel axis direction, and for example, a range of ±10° or ±15° with respect to the tunnel axis direction.

[0026] According to this configuration, since the first normal vector of the mesh plane substantially orthogonal to the tunnel axis direction is deleted, it is possible to exclude the artificial excavation surface substantially orthogonal to the tunnel axis direction and replace only the position and orientation of the faces other than the artificial excavation surface onto the two-dimensional plane.

[0027] Thereby, the face analysis system can not only efficiently color-code the sections of the section data, but also exclude the artificial excavation surface from the evaluation target, so that the evaluation of the face can be made easier.

[0028] Also, as an aspect of the present invention, there may be provided direction specifying means for specifying the direction of the second normal vector with a high appearance frequency, and the output means may be configured to output the section data in which the sections are color-coded with the predetermined color corresponding to the direction of the second normal vector specified by the direction specifying means.

[0029] The above direction specifying means refers to means for receiving a user's operation and specifying the direction of the second normal vector with a high appearance frequency, or means for automatically specifying the direction of the second normal vector with a high appearance frequency by means of a machine learning model or the like.

[0030] According to this configuration, since the sectional data color-coded with a predetermined color corresponding to the direction of the second normal vector with a high appearance frequency is output, for example, only the prominent cracks appearing on the face of the heading can be replaced on a two-dimensional plane. For this reason, the heading face analysis system can more easily evaluate the heading face.

[0031] Also, as an aspect of the present invention, the direction specifying means may be configured to receive a user's operation based on a distribution diagram of appearance frequencies and specify the direction of the second normal vector with a high appearance frequency. According to this configuration, since the user specifies the direction of the second normal vector while checking the distribution diagram of appearance frequencies, the sectional data desired by the user can be surely output.

[0032] Also, as an aspect of the present invention, a client terminal used by a user and a server connected to the client terminal via a communication line are provided. The client terminal includes video transmission means for transmitting video data obtained by photographing the heading face to the server, and display means for displaying the sectional data acquired from the server. The server includes the point group data acquisition means for acquiring the point group data from the video data, the three-dimensional model generation means, the first normal vector calculation means, the sectional data generation means, the second normal vector acquisition means, and the output means, and data transmission means for transmitting the color-coded sectional data to the client terminal.

[0033] The above communication line refers to a wired line such as an Internet line or a LAN line, or / and a wireless line. The above server refers to a server managed by a user, or a cloud server managed by a cloud service provider or the like.

[0034] According to this configuration, just by sending the video data of the face of the heading face to the server, a series of processes from the acquisition of the point cloud data to the output of the segmented data that has been color-segmented can be performed by the server. As a result, the heading face analysis system can reduce the load on the client terminal compared to the case where a series of processes are performed on the client terminal. Therefore, for example, a lightweight mobile terminal or the like can be used as the client terminal.

[0035] Note that the heading face analysis system may be configured such that an operation by the user is performed in response to a request from the server during a series of processes from the acquisition of the point cloud data to the output of the segmented data that has been color-segmented.

[0036] Furthermore, for example, by sending the segmented data to a client terminal different from the client terminal that sent the video data, the heading face analysis system can share the color-segmented data between, for example, the client terminal at the excavation site and the client terminal at a remote location away from the excavation site.

[0037] As a result, the heading face analysis system can not only reduce the burden on the operator by using a lightweight client terminal, but also enable multiple users to evaluate the state of the heading face.

Advantages of the Invention

[0038] According to the present invention, it is possible to provide a heading face analysis system and a heading face analysis method that facilitate the evaluation of the heading face using a three-dimensional model showing the shape of the heading face.

Brief Description of the Drawings

[0039]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Mode for Carrying Out the Invention

[0040] One embodiment of this invention will be described below with reference to the drawings. This embodiment will be described with reference to FIGS. 1 to 3 regarding a face analysis system 1 that analyzes the face F of a tunnel T and supports the evaluation of the face F by an operator M.

[0041] Note that FIG. 1 shows a schematic explanatory diagram for explaining the outline of the face analysis system 1, FIG. 2 shows a configuration diagram of the face analysis system 1, and FIG. 3 shows a block diagram of the face analysis system 1.

[0042] Also, the arrow Xi and the arrow Xo in the figure are in the tunnel axis direction. The arrow Xi in the figure indicates the excavation direction of the tunnel T (hereinafter referred to as the excavation direction Xi), and the arrow Xo indicates the portal direction opposite to the excavation direction Xi (hereinafter referred to as the portal direction Xo). Furthermore, the arrow Y in the figure indicates a direction that is substantially orthogonal to the excavation direction Xi in plan view (hereinafter referred to as the width direction Y).

[0043] First, as shown in FIGS. 1 and 2, the face analysis system 1 includes a portable terminal 10 with a camera that is used by the operator M as a user at the excavation site of the tunnel T, a management terminal 20 at a remote location away from the excavation site, and a server 30 connected to the portable terminal 10 and the management terminal 20 via a communication line 2.

[0044] Such a face analysis system 1 is configured such that, inside the tunnel T, the operator M uses the portable terminal 10 to photograph the face F and transmit it to the server 30, and after the server 30 analyzes the state of the face F, the analysis result is transmitted to the portable terminal 10 and the management terminal 20. Note that the face analysis system 1 is configured to be able to transmit the photographing information (video data 36 described later) of the face F from the portable terminal 10 and the management terminal 20 to the server 30.

[0045] Specifically, as shown in FIG. 3, the portable terminal 10 includes an operation display unit 11 that displays various information and accepts various operations of the operator M, a camera 12 that images a subject, a terminal storage unit 13 that stores various information, a line connection unit 14 that connects to the communication line 2, and a terminal control unit 15 that controls these operations.

[0046] Specifically, as shown in FIG. 2, the operation display unit 11 is composed of, for example, a touch panel display, and has a function of receiving various operations of the user and a function of displaying various information acquired from the server 30. The camera 12 has a function of imaging a subject as a moving image based on a control signal from the terminal control unit 15, and a function of outputting an imaging signal obtained by imaging the subject to the terminal control unit 15.

[0047] The terminal storage unit 13 is composed of, for example, a hard disk or a non-volatile memory, and has a function of writing and storing various information and a function of reading out various information. An analysis application 13a related to the evaluation support of the face F is stored in the terminal storage unit 13.

[0048] The line connection unit 14 is composed of, for example, a wireless communication module, and has a function of connecting to the communication line 2 and a function of transmitting and receiving various information via the communication line 2. The terminal control unit 15 is composed of hardware such as a CPU and a memory, and software such as a control program.

[0049] This terminal control unit 15 has a processing function related to the exchange of various information with the server 30, a processing function related to the exchange of various signals with the operation display unit 11, the camera 12, the terminal storage unit 13, and the line connection unit 14, and a function of controlling the operations of each unit connected via a predetermined bus.

[0050] The management terminal 20 is a terminal used by users working in a management office installed outside the tunnel T, for example, or workers at the excavation site of another tunnel T as users. As shown in FIG. 3, the management terminal 20 includes an operation reception unit 21 that receives various operations of the user, a display unit 22 that displays various information, a storage unit 23 that stores various information, a line connection unit 24 that connects to the communication line 2, and a control unit 25 that controls these operations.

[0051] Specifically, as shown in FIG. 2, the operation reception unit 21 is composed of, for example, a keyboard 21a, a mouse 21b, etc., and has a function of receiving an input operation by the user and a function of outputting information indicating the received input content to the control unit 25. As shown in FIG. 2, the display unit 22 is composed of, for example, a liquid crystal display, etc., and has a function of displaying various information according to a control signal from the control unit 25.

[0052] The storage unit 23 is composed of a hard disk or a non-volatile memory, etc., and has a function of writing and storing various information and a function of reading out various information. The same analysis application 23a as that of the portable terminal 10 is stored in this storage unit 23.

[0053] The line connection unit 24 is composed of, for example, a wired LAN board, etc., and has a function of connecting to the communication line 2 and a function of transmitting and receiving various information via the communication line 2. The control unit 25 is composed of hardware such as a CPU and a memory, and software such as a control program.

[0054] This control unit 25 has a processing function related to the exchange of various information with the server 30, a processing function related to the exchange of various signals with the operation reception unit 21, the display unit 22, the storage unit 23, and the line connection unit 24, and a function of controlling the operations of each unit connected via a predetermined bus.

[0055] As shown in FIG. 3, the server 30 includes a line connection unit 31 that connects to the communication line 2, a server storage unit 32 that stores various information, and a server control unit 33 that controls these operations. Specifically, the line connection unit 31 is composed of, for example, a wired LAN board, etc., and has a function of connecting to the communication line 2 and a function of transmitting and receiving various information via the communication line 2.

[0056] The server storage unit 32 is composed of a hard disk or a non-volatile memory, etc., and has a function of writing and storing various information and a function of reading out various information. As shown in FIG. 3, the server storage unit 32 stores an analysis program 34 for analyzing the state of the face F, project data 35 set and registered for each excavation site, and the like.

[0057] Furthermore, the server storage unit 32 registers, as project data 35, video data 36 of the face F acquired from the mobile terminal 10, a three-dimensional model 37 and distribution data 38 generated by the face analysis process described later, and the like.

[0058] More specifically, the three-dimensional model 37 is shape data showing the shape of the face F, and is polygon mesh data composed of a point group arranged according to the unevenness of the face F and triangular mesh planes having three adjacent points as vertices. The distribution data 38 is data showing the frequency distribution of the direction indicated by the second normal vector described later, and is configured to be outputtable as, for example, a Schmidt net.

[0059] The server control unit 33 is composed of hardware such as a CPU and a memory, and software such as a control program. This server control unit 33 has a processing function related to the exchange of various information with the mobile terminal 10 and the management terminal 20, a processing function related to the exchange of various signals with the line connection unit 31 and the server storage unit 32, and a function of controlling the operations of each unit connected via a predetermined bus.

[0060] Next, the processing operations in the face analysis system 1 having the above-described configuration will be described with reference to FIGS. 4 to 17. Note that FIG. 4 shows an explanatory diagram for explaining the photographing state of the face F, FIG. 5 shows a sequence diagram of the face analysis system 1, FIG. 6 shows a menu screen explanatory diagram for explaining the outline of the menu screen 200, and FIG. 7 shows a flowchart of the face analysis process.

[0061] Furthermore, FIG. 8 shows an explanatory diagram for explaining the relationship between the first normal vector V and the partition data 39A, FIG. 9 shows an explanatory diagram for explaining the outline of the partition data 39A, FIG. 10 shows an explanatory diagram for the distribution data 38 using the Schmidt net, and FIG. 11 shows a guidance screen explanatory diagram for explaining the outline of the guidance screen 210.

[0062] Furthermore, FIG. 12 shows an explanatory diagram for explaining the colored partition data 39B in which the first azimuth range and the fourth azimuth range are color-coded, and FIG. 13 shows an explanatory diagram for explaining the colored partition data 39B in which the second azimuth range and the third azimuth range are color-coded.

[0063] Also, FIG. 14 shows an explanatory diagram for explaining the colored partition data 39B in which the first azimuth range and the second azimuth range are color-coded, and FIG. 15 shows an explanatory diagram for explaining the colored partition data 39B in which the third azimuth range and the fourth azimuth range are color-coded. In addition, FIG. 16 shows an analysis result screen explanatory diagram for explaining the outline of the analysis result screen 220, and FIG. 17 shows an explanatory diagram for explaining the state in which the colored partition data 39B is displayed in time series.

[0064] First, as shown in FIG. 4, the operator M installs a pair of first markers 3 near both ends in the width direction Y on the face F in the tunnel T. Further, the operator M installs a second marker 4 at a predetermined interval in the width direction Y between the first markers 3. Note that the first marker 3 and the second marker 4 are, for example, AR (Augmented Reality) markers to which coordinate information indicating three-dimensional coordinates of an absolute coordinate system is assigned.

[0065] Thereafter, when the operator M operates the mobile terminal 10 to execute the analysis application 13a, the terminal control unit 15 of the mobile terminal 10 displays a menu screen 200 for guiding various operations on the operation display unit 11 as shown in FIG. 5 (step S101).

[0066] For example, on the menu screen 200, as shown in FIG. 6, there are displayed a title "Menu", a project name pre-entered, a shooting mode button 201 for shifting to the shooting mode of the face F, and a list acquisition button 202 for acquiring a list of analysis results from the server 30.

[0067] Furthermore, on the menu screen 200, in order to visualize the transition of the state of the discontinuity surface in the excavation direction Xi, there are displayed a time series button 203 for outputting a screen in which the color-coded section data 39B described later is juxtaposed in chronological order, an end button 204 for ending the analysis application 13a, and the like.

[0068] When the menu screen 200 is displayed on the operation display unit 11 of the mobile terminal 10, the operator M presses the shooting mode button 201 according to the guidance on the menu screen 200. When detecting the pressing of the shooting mode button 201 by the operator M, as shown in FIG. 5, the terminal control unit 15 starts the face shooting process (step S102) and displays on the operation display unit 11 a screen (not shown) for prompting the shooting of the face F.

[0069] At this time, the operator M stands at a position separated from the face F by a predetermined interval in the shaft mouth direction Xo and on one side in the width direction Y according to the guidance displayed on the operation display unit 11, and holds the camera 12 of the mobile terminal 10 facing the face F. In addition, the operator M adjusts the angle of view of the camera 12 so that the first marker 3 and the second marker 4 are displayed on the operation display unit 11 together with the face F.

[0070] After that, the operator M presses a shooting button (not shown) displayed on the operation display unit 11. When detecting the pressing of the shooting button by the operator M, based on the control signal from the terminal control unit 15, the camera 12 starts shooting a moving image of the subject and then outputs an imaging signal to the terminal control unit 15.

[0071] At this time, as shown in FIG. 4, the operator M photographs the face F while moving in the width direction Y substantially parallel to the face F. Then, when the movement in the width direction Y is completed, the operator M presses the photographing button again to finish photographing the face F. When detecting that the operator M presses the photographing button again, the terminal control unit 15 generates video data 36 associating the photographing date and time with the imaging signal and stores it in the terminal storage unit 13.

[0072] After that, when the operator M presses the transmission button displayed on the operation display unit 11, the terminal control unit 15 starts video transmission processing as shown in FIG. 5 (step S103) and transmits the video data 36 associated with the information indicating the project name to the server 30.

[0073] At this time, the server control unit 33 of the server 30 that has received the video data 36 from the mobile terminal 10 stores the video data 36 acquired from the mobile terminal 10 in the project data 35 corresponding to the project name.

[0074] Furthermore, the server control unit 33 starts a face F analysis process for analyzing the state of the face F based on the video data 36 as shown in FIG. 5 (step S104). When starting the face F analysis process, the server control unit 33 of the server 30 converts the video data 36 into point cloud data as shown in FIG. 7 (step S121).

[0075] Specifically, the server control unit 33 extracts a plurality of still images from the video data 36 and detects the shape of the face F based on the relative positions of the face F, the first marker 3, and the second marker 4 on the still image.

[0076] Furthermore, the server control unit 33 converts the video data 36 into point cloud data by arranging point clouds along the detected shape of the face F. At this time, the server control unit 33 arranges a point group based on the three-dimensional coordinates assigned to the first marker 3 and the second marker 4 so that the face F is located on the X-Z plane, for example, with the X-Z plane of the three-dimensional coordinates being a plane orthogonal to the excavation direction Xi.

[0077] When acquiring point group data from the video data 36, the server control unit 33 generates a three-dimensional model 37 showing the shape of the face F based on the point group data as shown in FIG. 7 (step S122). At this time, unnecessary point groups in the point group data may be deleted in advance before generating the three-dimensional model 37. After that, the server control unit 33 calculates a first normal vector, which is the normal vector on each mesh plane of the three-dimensional model 37 (step S123).

[0078] Specifically, as shown in FIG. 8(a), the server control unit 33 calculates position coordinates indicating the position of the center of gravity G on the mesh plane 37a based on the position coordinates of the three points P that are the vertices of the triangular mesh plane 37a among the point groups constituting the three-dimensional model 37.

[0079] After that, on the surface on the Xo side of the shaft direction in the three-dimensional model 37, the server control unit 33 calculates the first normal vector V along the normal direction orthogonal to the mesh plane 37a passing through the center of gravity G, and normalizes the magnitude to "1".

[0080] After calculating the first normal vector V for all the mesh planes 37a, as shown in FIG. 7, the server control unit 33 starts a deletion process of deleting the first normal vector V that is substantially parallel to the Y-axis direction of the three-dimensional coordinates among the calculated plurality of first normal vectors V (step S124).

[0081] Specifically, the server control unit 33 calculates the appearance frequency of the direction indicated by the first normal vector V, extracts the first normal vector V with a high appearance frequency, and then deletes the first normal vector V with the highest extracted appearance frequency.

[0082] At this time, the server control unit 33 sets a predetermined angular range obtained by adding a predetermined angle to the direction indicated by the first normal vector V with the highest appearance frequency as the deletion target range, and extracts the first normal vector V with the highest appearance frequency and the first normal vector V whose direction indicated by the first normal vector is the deletion target range as the first normal vector V with a high appearance frequency. Note that the deletion target range is an angular range considering the variation in the direction of the artificial excavation surface.

[0083] When the deletion process is completed, the server control unit 33 generates partition data 39A for partitioning the face F shown in the three-dimensional model 37 into partitions 39s of a predetermined size, as shown in FIG. 7 (step S125).

[0084] Specifically, as shown in FIG. 9, the server control unit 33 generates, on the three-dimensional model 37, lattice-shaped partition data 39A in which partitions 39s of, for example, 10 cm square are two-dimensionally arranged and overlapped, thereby partitioning the face F shown in the three-dimensional model 37.

[0085] In such a partition 39s of the partition data 39A, as shown in FIG. 8(b), there are cases where one center of gravity G of the mesh plane 37a in the three-dimensional model 37 is included, cases where a plurality of centers of gravity G of the mesh plane 37a are included, and cases where no center of gravity G of the mesh plane 37a is included.

[0086] After generating the partition data 39A, the server control unit 33 extracts one partition 39s of the partition data 39A, as shown in FIG. 7, and determines whether there are a plurality of centers of gravity G within the partition 39s (step S126).

[0087] Although detailed illustration is omitted, when there is no center of gravity G within the partition 39s, or when the first normal vector V is deleted in step S124 described above and there is no first normal vector V passing through the center of gravity G within the partition 39s, the server control unit 33 proceeds to step S130 described later and determines whether the processing from step S126 to step S129 for all partitions 39s has been completed.

[0088] When there are multiple center of gravity Gs within the section 39s (step S126: Yes), the server control unit 33 determines whether there are multiple first normal vectors V within the section 39s (step S127).

[0089] When there are multiple first normal vectors V within the section 39s (step S127: Yes), the server control unit 33 calculates the average value of the first normal vectors V within the section 39s (step S128). In other words, the server control unit 33 averages the orientation of the mesh plane 37a within the section 39s and calculates the normal vector of the mesh plane having the averaged orientation.

[0090] After that, the server control unit 33 normalizes the average value of the first normal vectors V within the section 39s to determine the second normal vector (step S129), and then assigns the vector information indicating the second normal vector to the section 39s and temporarily stores it. On the other hand, in step S126, when there is one center of gravity G within the section 39s (step S126: No), or in step S127, when there is one first normal vector V within the section 39s (step S127: No), the server control unit 33 determines the first normal vector V as the second normal vector within the section 39s as shown in FIG. 7 (step S129).

[0091] At this time, the server control unit 33 assigns the vector information indicating the second normal vector to the section 39s and temporarily stores it. After determining the second normal vector, the server control unit 33 determines whether the processing from step S126 to step S129 for all the sections 39s has been completed as shown in FIG. 7 (step S130).

[0092] When the processing for all the sections 39s has not been completed (step S130: No), the server control unit 33 returns the processing to step S126 and repeats the processing from step S126 to step S130 until the processing for all the sections 39s is completed.

[0093] On the other hand, when the processing for all the sections 39s is completed (step S130: Yes), the server control unit 33 stores the section data 39A with vector information added to each section 39s in the server storage unit 32, and then generates distribution data 38 indicating the frequency distribution of the vector information (step S131).

[0094] Specifically, the server control unit 33 calculates the appearance frequency in the direction indicated by the second normal vector based on the vector information added to each section 39s of the section data 39A, and then generates distribution data 38 that can be output as a Schmidt net of the appearance frequency distribution and stores it in the server storage unit 32. After generating the distribution data 38, the server control unit 33 transmits the distribution data 38 to the mobile terminal 10 (step S132), and then ends the face analysis process.

[0095] Note that, as shown in FIG. 10, the distribution data 38 is generated so as to be outputtable as a Schmidt net in which a first azimuth range, a second azimuth range, a third azimuth range, and a fourth azimuth range are defined with reference to the tunnel axis direction (excavation direction Xi and portal direction Xo).

[0096] Here, using the Schmidt net in the case where the excavation direction Xi shown in FIG. 10 is the north direction and the portal direction Xo is the south direction, the first azimuth range, the second azimuth range, the third azimuth range, and the fourth azimuth range will be described. The first azimuth range is set in the range of 90 degrees clockwise from the excavation direction Xi (north in the Schmidt net of FIG. 10).

[0097] Furthermore, the second azimuth range is set in the range of 90 degrees clockwise from the portal direction Xo (south in the Schmidt net of FIG. 10), the third azimuth range is set in the range of 90 degrees counterclockwise from the excavation direction Xi, and the fourth azimuth range is set in the range of 90 degrees counterclockwise from the portal direction Xo.

[0098] More specifically, the Schmidt net illustrates the frequency distribution of vector information in which the direction indicated by the second normal vector is upward and rightward in the first azimuth range, and illustrates the frequency distribution of vector information in which the direction indicated by the second normal vector is downward and rightward in the second azimuth range.

[0099] Furthermore, the Schmidt net illustrates the frequency distribution of vector information in which the direction indicated by the second normal vector is downward and leftward in the third azimuth range, and illustrates the frequency distribution of vector information in which the direction indicated by the second normal vector is upward and leftward in the fourth azimuth range.

[0100] In step S104 of FIG. 5, when the mobile terminal 10 receives the distribution data 38 transmitted by the server 30, the terminal control unit 15 of the mobile terminal 10 displays, on the operation display unit 11, a guidance screen 210 for guiding the color separation operation of the partition data 39A by the operator M as shown in FIG. 5 (step S105).

[0101] For example, as shown in FIG. 11, the guidance screen 210 displays a title "Color Separation Guidance", the project name and the shooting date and time of the video data 36, and a guidance message "Please select two locations with high occurrence frequencies".

[0102] Furthermore, on the guidance screen 210, the Schmidt net 211 of the distribution data 38 acquired from the server 30, a determination button 212 for determining the selection location, a return button 213 for returning to the menu screen 200, etc. are displayed.

[0103] At this time, according to the guidance on the guidance screen 210, the operator M selects and presses one location with a high occurrence frequency in each of two different azimuth ranges among the four azimuth ranges (the first azimuth range, the second azimuth range, the third azimuth range, and the fourth azimuth range), and then presses the determination button 212. For example, the operator M presses the locations with high occurrence frequencies in the first azimuth range and the fourth azimuth range (the circles in FIG. 10), and then presses the determination button 212.

[0104] When the terminal control unit 15 detects that the operator M presses the determination button 212, as shown in FIG. 5, the terminal control unit 15 transmits the pressing position coordinates on the Schmitt net 211 to the server 30 as selection pressing information (step S106).

[0105] When the server 30 acquires the selection pressing information transmitted from the mobile terminal 10, as shown in FIG. 5, the server control unit 33 of the server 30 generates color-coded partition data 39B obtained by color-coding the partitions 39s of the partition data 39A based on the selection pressing information (step S107).

[0106] Specifically, the server control unit 33 sets the highest appearance frequency among the appearance frequencies corresponding to the pressing position on the Schmitt net 211 indicated by the pressing position coordinates as the highest appearance frequency, and extracts the third or fourth highest appearance frequencies counted from the highest appearance frequency within a predetermined range centered on the pressing position from the distribution data 38.

[0107] Furthermore, the server control unit 33 extracts the direction of the second normal vector corresponding to the extracted appearance frequency from the distribution data 38, and detects the partition 39s of the partition data 39A to which vector information indicating the direction of the extracted second normal vector is assigned.

[0108] Thereafter, the server control unit 33 color-codes the partition data 39A by assigning a color attribute indicating a predetermined color set in advance for each of the four azimuth ranges to the detected partition 39s. Note that it is desirable that the predetermined colors are different for each of the four azimuth ranges, but they may all be the same color.

[0109] For example, in step S105 of FIG. 5, when the portions with high appearance frequencies in the first azimuth range and the fourth azimuth range (circles in FIG. 10) are pressed, as shown in FIG. 12, the color-coded partition data 39B is in a state where the partition 39s corresponding to the pressing position on the Schmitt net 211 are color-coded with a predetermined color indicating the first azimuth range and a predetermined color indicating the fourth azimuth range.

[0110] In this case, among the direction components of the first azimuth range and the fourth azimuth range, since the vertical directions are both upward, the colored section data 39B shows the upward-facing surfaces (e.g., the dominant cracks of the flow marks) among the surfaces other than the artificial excavation surface.

[0111] Also, in step S105 of FIG. 5, when the locations with high occurrence frequencies in the second azimuth range and the third azimuth range (the circled marks in FIG. 10) are pressed, as shown in FIG. 13, the colored section data 39B is in a state where the section 39s corresponding to the pressed position on the Schmidt net 211 is color-coded with a predetermined color indicating the second azimuth range and a predetermined color indicating the third azimuth range.

[0112] In this case, among the direction components of the second azimuth range and the third azimuth range, since the vertical directions are both downward, the colored section data 39B shows the downward-facing surfaces (e.g., the dominant cracks of the cross joints) among the surfaces other than the artificial excavation surface.

[0113] Also, in step S105 of FIG. 5, when the locations with high occurrence frequencies in the first azimuth range and the second azimuth range (the circled marks in FIG. 10) are pressed, as shown in FIG. 14, the colored section data 39B is in a state where the section 39s corresponding to the pressed position on the Schmidt net 211 is color-coded with a predetermined color indicating the first azimuth range and a predetermined color indicating the second azimuth range.

[0114] In this case, among the direction components of the first azimuth range and the second azimuth range, since the horizontal directions are both rightward, the colored section data 39B shows the rightward-facing surfaces (e.g., the dominant cracks in the rightward direction) among the surfaces other than the artificial excavation surface.

[0115] Also, in step S105 of FIG. 5, when the locations with high occurrence frequencies in the third azimuth range and the fourth azimuth range (the circled marks in FIG. 10) are pressed, as shown in FIG. 15, the colored section data 39B is in a state where the section 39s corresponding to the pressed position on the Schmidt net 211 is color-coded with a predetermined color indicating the third azimuth range and a predetermined color indicating the fourth azimuth range.

[0116] In this case, among the directional components of the third azimuth range and the fourth azimuth range, since both the left-right directions are leftward, the colored section data 39B shows a leftward surface (for example, a dominant leftward crack) among the surfaces other than the artificial excavation surface.

[0117] In the above-mentioned step S107, when the colored section data 39B is generated, as shown in FIG. 5, the server control unit 33 transmits the colored section data 39B to the mobile terminal 10 (step S108).

[0118] On the other hand, the terminal control unit 15 of the mobile terminal 10 that has acquired the colored section data 39B from the server 30 displays, on the operation display unit 11, an analysis result screen 220 that presents the colored section data 39B to the operator M as an analysis result (step S109).

[0119] For example, as shown in FIG. 16, on the analysis result screen 220, a title "Analysis Result", the project name, and the shooting date and time of the video data 36 are displayed. Below that, the colored section data 39B acquired from the server 30, a return button 221 for returning to the guidance screen 210, an end button 222 for ending the analysis application 13a, etc. are displayed.

[0120] When the analysis result screen 220 is displayed on the mobile terminal 10, the operator M at the excavation site estimates the locations where skin loss is predicted and the discontinuous surfaces that have appeared on the face F based on the displayed colored section data 39B. In this way, the face analysis system 1 analyzes the face F of the tunnel T and supports the operator M in evaluating the face F.

[0121] Also, although detailed illustration is omitted, in step S101 of FIG. 5, when the list acquisition button 202 (see FIG. 6) on the menu screen 200 is pressed by the operator M, the terminal control unit 15 of the mobile terminal 10 outputs to the operation display unit 11 a screen on which a list of the shooting dates and times acquired from the server 30 and a message prompting the selection of the shooting date and time are displayed.

[0122] After that, when an operator M selects one from the list of shooting dates and times, the terminal control unit 15 of the mobile terminal 10 requests the server 30 to transmit the distribution data 38, and proceeds to step S105 in FIG. 5, and outputs a guidance screen 210 on which the acquired distribution data 38 is displayed to the operation display unit 11.

[0123] Then, the face analysis system 1 enables the viewing of the colored section data 39B at an arbitrary timing by the mobile terminal 10 performing the processes of steps S106 and S109 described above and the server 30 performing the processes of steps S107 and S108 described above.

[0124] Also, in step S101 of FIG. 5, when the time series button 203 (see FIG. 6) on the menu screen 200 is pressed by the operator M, the server control unit 33 generates a plurality of colored section data 39B based on a plurality of distribution data 38 with different dates and times.

[0125] Then, as shown in FIG. 17, the face analysis system 1 enables the visualization of the transition of the discontinuity plane in the excavation direction Xi by juxtaposing the colored section data 39B generated by the server control unit 33 in chronological order and outputting it to the mobile terminal 10 and the management terminal 20.

[0126] Also, in the management terminal 20 away from the tunnel T, when the analysis application 23a is executed and the user of the management terminal 20 presses the list acquisition button 202 on the menu screen 200, the control unit 25 of the management terminal 20 outputs to the display unit 22 a screen on which a list of shooting dates and times acquired from the server 30 and a message prompting the selection of the shooting date and time are displayed.

[0127] Then, the face analysis system 1 enables the viewing of the colored section data 39B at a remote location by the control unit 25 of the management terminal 20 performing processing in the same manner as when the list acquisition button 202 on the menu screen 200 is pressed on the mobile terminal 10.

[0128] In this way, the face analysis system 1 of the present embodiment enables the colored section data 39B indicating the state of the face F, which is generated based on the moving image data 36 obtained by photographing the face F, to be displayed on both the portable terminal 10 at the excavation site and the management terminal 20 at a remote location at any timing.

[0129] As described above, the face analysis system 1 of the present embodiment is a system that analyzes the face F of the tunnel T and supports the evaluation of the face F. This face analysis system 1 includes a point cloud data acquisition means (server control unit 33) for acquiring point cloud data indicating the shape of the face F, and a three-dimensional model generation means (server control unit 33) for generating a three-dimensional model 37 composed of triangular mesh planes 37a having adjacent points of the point cloud data as vertices.

[0130] Furthermore, the face analysis system 1 includes a first normal vector calculation means (server control unit 33) for calculating a first normal vector V along the normal direction passing through the centroid G of the mesh plane 37a, and a section data generation means (server control unit 33) for generating section data 39A in which sections 39s of a predetermined size are two-dimensionally arranged on the three-dimensional model 37.

[0131] In addition, the face analysis system 1 includes a second normal vector acquisition means (server control unit 33) for acquiring, for each section 39s, one normal vector obtained based on the first normal vector V included in the section 39s of the section data 39A as the second normal vector in the section 39s.

[0132] And the face analysis system 1 includes an output means (server control unit 33) for outputting colored section data 39B in which the sections 39s are color-coded with a predetermined color corresponding to the direction indicated by the second normal vector.

[0133] Also, the face analysis method in the present embodiment is a method that analyzes the face F of the tunnel T and supports the evaluation of the face F. This face analysis method includes a point cloud data acquisition step (step S121) in which point cloud data acquisition means acquires point cloud data indicating the shape of the face F, and a three-dimensional model generation step (step S122) in which three-dimensional model generation means generates a three-dimensional model 37 composed of triangular mesh planes 37a with adjacent points of the point cloud data as vertices.

[0134] Furthermore, the face analysis method includes a first normal vector calculation step (step S123) in which first normal vector calculation means calculates a first normal vector V along the normal direction passing through the centroid G of the mesh plane 37a, and a section data generation step (step S125) in which section data generation means generates section data 39A in which sections 39s of a predetermined size are two-dimensionally arranged on the three-dimensional model 37.

[0135] In addition, the face analysis method includes a second normal vector acquisition step (steps S126 to S129) in which second normal vector acquisition means acquires, for each section 39s, one normal vector obtained based on the first normal vector V included in the section 39s of the section data 39A as the second normal vector in the section 39s.

[0136] Then, the face analysis method performs an output step (step S107) in which output means outputs colored section data 39B in which the sections 39s are colored with a predetermined color corresponding to the direction indicated by the second normal vector.

[0137] According to this configuration, the position and orientation of the artificial excavation surface in the three-dimensional model 37 and the position and orientation of the surfaces other than the artificial excavation surface (for example, the discontinuous surface that appears on the face F) can be replaced on a two-dimensional plane.

[0138] Therefore, for example, when the mesh plane 37a of the three-dimensional model 37 is colored, it is necessary to observe the three-dimensional model 37 from multiple directions to evaluate the state of the face F, whereas the face analysis system 1 can enable the evaluation of the face F without observing from multiple directions.

[0139] Furthermore, through the cooperation of the section data generation means and the second normal vector acquisition means, the face analysis system 1 can acquire the second normal vector by regarding it as the normal vector of one mesh plane within the section 39s. As a result, the face analysis system 1 can output the colored section data 39B that has been color-coded while suppressing the influence of fine unevenness, thus enabling the state of the face F to be clarified.

[0140] Therefore, the face analysis system 1 and the face analysis method using the same can facilitate the evaluation of the face F using the three-dimensional model 37 as compared to the case of evaluating the face F using the three-dimensional model 37 in which the mesh planes 37a are color-coded.

[0141] Also, when there are multiple first normal vectors V within the section 39s of the section data 39A, the second normal vector acquisition means (server control unit 33) is configured to acquire the average value of the multiple first normal vectors V as the second normal vector. According to this configuration, it is possible to output the colored section data 39B that has been color-coded while reliably suppressing the influence of fine unevenness, thus enabling the state of the face F to be made clearer.

[0142] Moreover, since the point cloud data acquisition means is configured to acquire point cloud data from the video data 36 obtained by photographing the face F, the face analysis system 1 can acquire point cloud data from the video data 36 obtained by photographing the face F with the mobile terminal 10 equipped with a camera.

[0143] As a result, the face analysis system 1 can acquire point cloud data without installing a three-dimensional laser scanner or the like inside the tunnel T, thus reducing the burden on the worker M at the excavation site.

[0144] In addition, the face analysis system 1 includes a first normal vector extraction means (server control unit 33) that extracts the first normal vector V with a high frequency of occurrence, and a first normal vector deletion means (server control unit 33) that deletes the first normal vector V extracted by the first normal vector extraction means.

[0145] According to this configuration, by deleting the first normal vector V with a high appearance frequency, it is possible to exclude the replacement of the artificial excavation surface with a high appearance frequency onto the two-dimensional plane. In other words, the face analysis system 1 can replace only the position and orientation of the faces other than the artificial excavation surface onto the two-dimensional plane.

[0146] As a result, the face analysis system 1 can not only efficiently color-code the sections 39s of the sectional data 39A, but also exclude the artificial excavation surface from the evaluation target, so that the evaluation of the face F can be made easier.

[0147] In addition, the face analysis system 1 includes a direction specifying means (operation display unit 11 where the Schmidt net 211 is displayed) for specifying the direction of the second normal vector with a high appearance frequency. And the output means (server control unit 33) is configured to output the color-coded sectional data 39B in which the sections 39s are color-coded with a predetermined color corresponding to the direction of the second normal vector specified by the direction specifying means.

[0148] According to this configuration, since the color-coded sectional data 39B color-coded with a predetermined color corresponding to the direction indicated by the second normal vector with a high appearance frequency is output, for example, only the dominant cracks that have appeared on the face F can be replaced onto the two-dimensional plane. For this reason, the face analysis system 1 can make the evaluation of the face F easier.

[0149] In addition, the direction specifying means (operation display unit 11) is configured to receive the operation of the operator M based on the Schmidt net 211 and specify the direction of the second normal vector with a high appearance frequency. According to this configuration, since the user specifies the direction of the second normal vector while checking the Schmidt net 211, the user can surely output the color-coded sectional data 39B desired by the user.

[0150] In addition, the face analysis system 1 includes a mobile terminal 10 used by a user and a management terminal 20, and a server 30 connected to the mobile terminal 10 and the management terminal 20 via a communication line 2.

[0151] Furthermore, the mobile terminal 10 includes a video transmission means (terminal control unit 15) for transmitting video data 36 obtained by photographing the face F to the server 30, and a display means (operation display unit 11) for displaying the colored section data 39B obtained from the server 30.

[0152] On the other hand, the server 30 includes a point cloud data acquisition means for acquiring point cloud data from the video data 36, a three-dimensional model generation means, a first normal vector calculation means, a section data generation means, a second normal vector acquisition means and an output means, and a data transmission means (server control unit 33) for transmitting the color-separated colored section data 39B to the mobile terminal 10 and the management terminal 20.

[0153] According to this configuration, by simply transmitting the video data 36 obtained by photographing the face F to the server 30, the server 30 can perform a series of processes from the acquisition of the point cloud data to the output of the color-separated colored section data 39B.

[0154] Thereby, the face analysis system 1 can reduce the load on the client terminal compared to the case where a series of processes are performed on the client terminal, so that a lightweight mobile terminal 10 can be used as the client terminal.

[0155] Furthermore, by transmitting the colored section data 39B to a management terminal 20 different from the mobile terminal 10 that has transmitted the video data 36, the face analysis system 1 can share the color-separated colored section data 39B between the mobile terminal 10 at the excavation site and the management terminal 20 at a remote location away from the excavation site.

[0156] Thereby, the face analysis system 1 can not only reduce the burden on the operator M by a lightweight client terminal, but also enable a plurality of users to evaluate the state of the face F.

[0157] Also, when the time series button 203 on the menu screen 200 is pressed by the operator M, in order to output the colored section data 39B juxtaposed in chronological order to the mobile terminal 10 and the management terminal 20, the face analysis system 1 can facilitate the calculation of the strike and dip of the discontinuous surface and the estimation of the discontinuous surface on the excavation direction Xi side rather than the face F.

[0158] In the correspondence between the configuration of this invention and the above-described embodiment, The user of this invention corresponds to the operator M of the embodiment, Similarly hereinafter, The point cloud data acquisition means, the three-dimensional model generation means, the first normal vector calculation means, the section data generation means, the second normal vector acquisition means, the output means, the first normal vector extraction means, the first normal vector deletion means, and the data transmission means correspond to the server control unit 33, The color-separated section data corresponds to the colored section data 39B, The direction identification means and the display means correspond to the operation display unit 11, The distribution diagram of the appearance frequency corresponds to the Schmidt net 211, The client terminal corresponds to the mobile terminal 10, The video transmission means corresponds to the terminal control unit 15, The point cloud data acquisition step corresponds to step S121 in FIG. 7, The three-dimensional model generation step corresponds to step S122 in FIG. 7, The first normal vector calculation step corresponds to step S123 in FIG. 7, The section data generation step corresponds to step S125 in FIG. 7, The second normal vector acquisition step corresponds to steps S126 to S129 in FIG. 7, The output step corresponds to step S107 in FIG. 5, This invention is not limited to only the configuration of the above-described embodiment, and many embodiments can be obtained.

[0159] For example, in the above-described embodiment, the client terminal used by the operator M at the excavation site is the mobile terminal 10, but it is not limited thereto, and a tablet terminal or a notebook computer may be used as the client terminal.

[0160] Also, although the video of the face F is taken with the mobile terminal 10 equipped with a camera, it is not limited thereto, and as long as video data 36 can be output, a tablet terminal with a camera, a notebook computer with a camera, a digital camera, a video camera, or a wearable camera may be used. Also, the menu screen 200 in FIG. 6, the guidance screen 210 in FIG. 11, and the analysis result screen 220 in FIG. 16 are each an example and are not limited thereto, and an appropriate configuration may be used.

[0161] Also, when shooting the video of the face F, the first marker 3 and the second marker 4 are arranged near the face F, but it is not limited thereto. A pair of red-and-white poles may be installed near both ends in the width direction Y of the face F, and markers may be installed at predetermined intervals in the width direction Y between the red-and-white poles.

[0162] Also, although the point cloud data is generated based on the video data 36, it is not limited thereto, and the point cloud data based on the measurement data measured by the three-dimensional laser scanner may be used. Also, in step S103 of FIG. 5, the mobile terminal 10 that has shot the face F transmits the video data 36 to the server 30, but it is not limited thereto, and the video data 36 may be transmitted to the server 30 from a terminal different from the mobile terminal 10 that has shot the face F (for example, the management terminal 20).

[0163] Also, in step S121 of FIG. 7, the server 30 generates and acquires the point cloud data, but it is not limited thereto, and a terminal or a server different from the server 30 generates the point cloud data based on the video data, and the server 30 acquires the generated point cloud data via the communication line 2 and performs the subsequent processing.

[0164] Also, although not described in detail in the above-described embodiments, the three-dimensional model 37 generated in step S122 of FIG. 7 may be generated by the server 30 based on the point cloud data from which unnecessary point clouds have been automatically deleted.

[0165] Alternatively, the three-dimensional model 37 may be generated based on the point cloud data from which unnecessary point clouds have been deleted by the operation of the user. In this case, for example, the point cloud data generated by the server 30 is acquired by the management terminal 20, and in the management terminal 20 that has acquired the point cloud data, a new point cloud data is generated by accepting the operation of the user and deleting and arranging the unnecessary point clouds. Then, the management terminal 20 transmits the new point cloud data to the server 30. And the server 30 generates the three-dimensional model 37 based on the new point cloud data in step S122.

[0166] Also, although the first normal vector V is normalized in step S123 of FIG. 7 and the second normal vector is normalized in step S129, the present invention is not limited to this, and the first normal vector V and the second normal vector may be used in the subsequent processing without being normalized respectively.

[0167] Also, in step S124 of FIG. 7, the first normal vector V with the highest appearance frequency and the first normal vector V whose direction indicated by the first normal vector is the deletion target range are extracted as the first normal vector V with a high appearance frequency, but the present invention is not limited to this. For example, the first normal vector with the highest appearance frequency may be extracted as the first normal vector V with a high appearance frequency.

[0168] Alternatively, the first normal vector V with the appearance frequency from the highest appearance frequency to the third or fourth may be extracted as the first normal vector V with a high appearance frequency. Even in this case, similar to the above-described embodiments, since the artificial excavation surface is excluded and only the position and orientation of the surfaces other than the artificial excavation surface can be replaced on the two-dimensional plane, the evaluation of the face F can be made easier.

[0169] Further, the deletion target range of the first normal vector V in step S124 of FIG. 7 may be configured to be arbitrarily changeable by the user. Also, in step S124 of FIG. 7, the first normal vector V with a high appearance frequency was deleted, but it is not limited to this, and the generation of the partition data 39A may be performed without deleting the first normal vector V with a high appearance frequency.

[0170] Alternatively, after extracting the first normal vector V whose direction indicated by the first normal vector V is within a predetermined angle range with respect to the Y-axis direction (the heading direction Xo), the extracted first normal vector V may be deleted.

[0171] Note that the predetermined angle range is an angle range that is substantially orthogonal to the heading direction Xo, for example, a range of ±10° or ±15° with respect to the heading direction Xo. Further, the predetermined angle range may be configured to be arbitrarily changeable by the user.

[0172] According to this configuration, since the first normal vector V of the mesh plane 37a that is substantially orthogonal to the heading direction Xo is deleted, an artificial excavation surface that is substantially orthogonal to the heading direction Xo is excluded, and only the position and orientation of the surface other than the artificial excavation surface can be replaced on the two-dimensional plane.

[0173] Thereby, the face analysis system 1 can not only efficiently color-code the partitions 39s of the partition data 39A, but also exclude the artificial excavation surface from the evaluation target, so that the evaluation of the face F can be made easier.

[0174] Also, in step S125 of FIG. 7, it was generated on the three-dimensional model 37 so as to overlap the partition data 39A, but it is not limited to this, and the three-dimensional model 37 may be directly partitioned to generate the partition data 39A.

[0175] Also, the size of the partition 39s of the partition data 39A was set to 10 cm square, but the size of the partition 39s may be an appropriate size. Further, the size of the partition 39s may be configured to be arbitrarily changeable by the user. Further, the section 39s of the section data 39A is not limited to the shape of the above-described embodiment, and may be a polygonal section 39s such as a hexagon as long as it can be two-dimensionally arranged.

[0176] Also, in step S128 of FIG. 7, when there are a plurality of first normal vectors V in the section 39s, the average value of the first normal vectors V in the section 39s is used as the second normal vector. However, the present invention is not limited to this, and one normal vector may be obtained by an appropriate method based on the first normal vectors V included in the section 39s, and the obtained one normal vector may be used as the second normal vector.

[0177] For example, the area ratio of the mesh plane 37a in the section 39s may be calculated, and the first normal vector V passing through the centroid G of the mesh plane 37a with a relatively large area ratio may be used as the second normal vector. Also, although the distribution data 38 is output as a Schmidt net, the present invention is not limited to this, and for example, the distribution data 38 may be output as a two-dimensional array table.

[0178] Also, in step S105 of FIG. 5, the direction of the second normal vector with a high appearance frequency is specified by the operator M pressing the selection button. However, the present invention is not limited to this, and a configuration may be adopted in which the direction of the second normal vector with a high appearance frequency is automatically specified. For example, the direction of the second normal vector with a high appearance frequency may be automatically specified by the cooperation of the distribution data 38 and the machine learning model.

[0179] Also, in step S105 of FIG. 5, the section 39s to which the color attribute is to be assigned is determined by causing the operator M to select and press the Schmidt net 211 on the guidance screen 210. However, the present invention is not limited to this, and the section 39s to which the color attribute is to be assigned may be determined by causing the operator M to select and press a button for selecting the orientation of the illustrated surface.

[0180] In this case, as shown in FIG. 18, which is an explanatory diagram for explaining the selection screen 230 in another embodiment, a selection screen 230 for selecting the orientation of the illustrated surface is displayed on the operation display unit 11 of the mobile terminal 10 to prompt the operator M to select and press.

[0181] On this selection screen 230, as shown in FIG. 18, there are displayed a flow eye button 231 for selecting an upward-facing surface, a notch button 232 for selecting a downward-facing surface, a left-facing button 233 for selecting a left-facing surface, and a right-facing button 234 for selecting a right-facing surface. Note that the selection screen 230 in FIG. 18 is an example, and it may have an appropriate configuration.

[0182] Then, the server control unit 33 of the server 30 outputs colored partition data 39B in which partitions 39s corresponding to the orientation of the discontinuous surface selected by the user are color-coded. For example, when the flow eye button 231 is pressed, the server control unit 33 outputs the colored partition data 39B shown in FIG. 12. Thereby, the cut-off surface analysis system 1 can achieve the same effects as those of the above-described embodiment.

[0183] Also, in step S107 of FIG. 5, within a predetermined range centered on the pressing position on the Schmitt net 211, the third or fourth highest occurrence frequencies were extracted starting from the highest occurrence frequency. However, the range of the predetermined range and the number of occurrence frequencies to be extracted from the highest occurrence frequency may be arbitrarily changed.

[0184] Also, colored partition data 39B in which partitions 39s with high occurrence frequency vector information are color-coded was generated. However, the present invention is not limited to this, and colored partition data 39B in which partitions 39s with vector information regardless of the occurrence frequency are color-coded may be generated.

[0185] Also, the colored partition data 39B in which the partitions 39s corresponding to the pressing position on the Schmitt net 211 pressed by the operator M are color-coded was displayed on the mobile terminal 10 and the management terminal 20. However, the present invention is not limited to this. For example, as shown in FIG. 19, which shows an explanatory diagram for explaining the colored partition data 39B in another embodiment, colored partition data 39B in which the partitions 39s corresponding to the locations with high occurrence frequencies in four azimuth ranges are color-coded with different colors may be further displayed.

[0186] Also, although the server 30 is configured to perform the processes of step S104 in FIG. 5 (steps S121 to S132 in FIG. 7), step S107, and step S108, the present invention is not limited to this. For example, a cross-section analysis system that performs the processes of step S104, step S107, and step S108 on the mobile terminal 10 may be used.

[0187] Alternatively, a cross-section analysis system may be used in which the server 30 performs the processes of step S121 of acquiring point cloud data, step S122 of generating a three-dimensional model 37, step S107 of outputting colored partition data 39B, and step S108 of transmitting the colored partition data 39B, and a terminal different from the mobile terminal 10 and the server 30 performs the processes of steps S123 to S132 in FIG. 7.

[0188] Also, although the above-described processes are started by executing the analysis application 13a stored in the terminal storage unit 13 of the mobile terminal 10, the present invention is not limited to this, and a so-called web application that starts the above-described processes by accessing a predetermined URL using a web browser may be used.

[0189] Also, the processing flow in the above-described embodiment is merely an example and is not limited thereto. For example, after determining the second normal vector in step S129 in FIG. 7, a color attribute indicating a predetermined color corresponding to the direction of the second normal vector may be assigned to the partition 39s to generate the colored partition data 39B.

[0190] At this time, the color attribute is assigned to the partition 39s in a state where the visualization and non-visualization of a predetermined color can be switched, and in subsequent processes, the colored partition data 39B in which the desired partition 39s are color-coded is output by appropriately visualizing and non-visualizing the color attribute.

Explanation of Reference Numerals

[0191] 1... Cross-section analysis system 2... Communication line 10... Mobile terminal 11... Operation display unit 15…Terminal control unit 30…Server 33…Server control unit 36…Video data 37…Three-dimensional model 37a…Mesh plane 39A…Partition data 39B…Colored partition data 39s…Partition 211…Schmitt net F…Face G…Center of gravity M…Operator T…Tunnel V…First normal vector Xo…Shaft direction

Claims

1. A face analysis system for analyzing the face of a tunnel and assisting in the evaluation of the face, comprising: point cloud data acquisition means for acquiring point cloud data indicating the shape of the face; three-dimensional model generation means for generating a three-dimensional model composed of triangular mesh planes with adjacent points of the point cloud data as vertices; first normal vector calculation means for calculating a first normal vector along the normal direction passing through the centroid of the mesh plane; section data generation means for generating section data in which sections of a predetermined size are two-dimensionally arranged on the three-dimensional model; second normal vector acquisition means for acquiring, for each section, one normal vector obtained based on the first normal vector included in the section of the section data as the second normal vector in the section; output means for outputting the section data in which the sections are color-coded with a predetermined color corresponding to the direction indicated by the second normal vector. Face analysis system.

2. The second normal vector acquisition means is configured to, when there are a plurality of first normal vectors in the section of the section data, acquire the average value of the plurality of first normal vectors as the second normal vector. The face analysis system according to claim 1.

3. The point cloud data acquisition means is configured to acquire the point cloud data from video data obtained by photographing the face. The face analysis system according to claim 1.

4. first normal vector extraction means for extracting the first normal vector with a high appearance frequency; first normal vector deletion means for deleting the first normal vector extracted by the first normal vector extraction means. The face analysis system according to claim 1.

5. first normal vector extraction means for extracting the first normal vector whose direction indicated by the first normal vector is within a predetermined angle range with respect to the tunnel axis direction; first normal vector deletion means for deleting the first normal vector extracted by the first normal vector extraction means. The face analysis system according to claim 1.

6. direction identification means for identifying the direction of the second normal vector with a high appearance frequency is provided, The output means is configured to output the section data in which the sections are color-coded with the predetermined color corresponding to the direction of the second normal vector identified by the direction identification means. The face analysis system according to claim 1.

7. The direction identification means Based on the distribution diagram of the appearance frequency, it is configured to receive the user's operation and identify the direction of the second normal vector with a high appearance frequency. The face analysis system according to claim 6.

8. A client terminal used by the user, and a server connected to the client terminal via a communication line, wherein the client terminal is provided with video transmission means for transmitting video data of the face to the server, and display means for displaying the section data obtained from the server, wherein the server is provided with the point cloud data acquisition means for acquiring the point cloud data from the video data, the three-dimensional model generation means, the first normal vector calculation means, the section data generation means, the second normal vector acquisition means, and the output means, and data transmission means for transmitting the color-coded section data to the client terminal. The face analysis system according to claim 1.

9. A face analysis method for analyzing the face of a tunnel and assisting in the evaluation of the face, comprising: a point cloud data acquisition step in which point cloud data acquisition means acquires point cloud data indicating the shape of the face; a three-dimensional model generation step in which three-dimensional model generation means generates a three-dimensional model composed of triangular mesh planes with adjacent points of the point cloud data as vertices; a first normal vector calculation step in which first normal vector calculation means calculates a first normal vector along the normal direction passing through the centroid of the mesh plane; a section data generation step in which section data generation means generates section data in which sections of a predetermined size are two-dimensionally arranged on the three-dimensional model; a second normal vector acquisition step in which second normal vector acquisition means acquires, for each section, one normal vector obtained based on the first normal vectors included in the section of the section data as the second normal vector in the section; and an output step in which output means outputs the section data in which the sections are color-coded with a predetermined color corresponding to the direction indicated by the second normal vector. Face analysis method.

Images