Apparatus, method and program for extracting linear objects

Abstract

The purpose of the present disclosure is to enable extraction of a point cloud of a wire-like object for which the point cloud is sparse, e.g., a cable.　The present disclosure is a device and a method for extracting a point cloud for a wire-like object such as a cable 92 by acquiring point cloud data representing three-dimensional coordinates and a movement trajectory at the time when the point cloud data is measured, and repeatedly counting a number of point clouds on the movement trajectory within a fixed distance DN from the point cloud.

Description

[Technical field]

[0001] The present disclosure relates to an apparatus, a method, and a program for creating a three-dimensional model. [Background technology]

[0002] A technology has been developed that uses an on-board 3D laser scanner (Mobile Mapping System: MMS) to measure point cloud data of outdoor structures and create a 3D model of the outdoor structures from the point cloud data (see, for example, Patent Documents 1 and 2).

[0003] In the technology of Patent Document 1, the point cloud acquired during one rotation of the laser scanner is made into a cluster called a scan line, and then adjacent clusters are detected to be in a catenary shape, thereby creating a three-dimensional model of the overhead cable. However, this technology has a problem in that it is difficult to create a scan line when point cloud data cannot be acquired sufficiently. Therefore, in the technology of Patent Document 1, when the number of point cloud outputs from the laser scanner is small or when point cloud data cannot be acquired sufficiently due to the influence of obstacles, etc., there is a problem in that the point cloud of the cable part cannot be extracted.

[0004] In addition, the technology of Patent Document 2 detects an overhead cable from a point cloud using travel trajectory information. However, this technology only determines the height of the cable above ground from the detected point cloud, and is unable to extract a point cloud of the cable portion. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent No. 6531051 [Patent Document 2] WO2021 / 038647 Summary of the Invention [Problem to be solved by the invention]

[0006] An object of the present disclosure is to enable extraction of a point cloud of a linear object, such as a cable, whose point cloud is sparse. [Means for solving the problem]

[0007] The apparatus and method of the present disclosure include: Acquire point cloud data representing three-dimensional coordinates and a movement trajectory when measuring the point cloud data, The point cloud of a linear object is extracted by repeatedly counting the number of point clouds that are within a certain distance from the point cloud on the movement trajectory. Effect of the Invention

[0008] According to the present disclosure, it is possible to extract a point cloud of a linear object such as a cable, whose point cloud is sparse. [Brief description of the drawings]

[0009] [Figure 1] 1 shows an overview of a system according to the present disclosure. [Diagram 2] An example of point cloud data measured with MMS is shown below. [Diagram 3] 1 shows an outline of a method for extracting linear objects according to the present disclosure. [Figure 4] A specific example of the first point group N will be shown. [Diagram 5] FIG. 11 is an explanatory diagram of a process for creating a point cloud group of cable candidates. [Figure 6] 11 is an explanatory diagram of a process of dividing a point cloud group of cable candidates into point cloud groups for each cable. FIG. [Figure 7] FIG. 11 is an explanatory diagram of a process for obtaining an approximation line. [Figure 8] 1 illustrates an example system configuration of the present disclosure. [Figure 9] 13 shows an example of a process performed by a point cloud determination unit. [Figure 10] An example of a cable model created from the point cloud N in FIG. 4 is shown. [Figure 11] 13 shows an example of a process performed by a point cloud determination unit. [Figure 12]11 is an explanatory diagram of a process for calculating a road width from the coordinates of a travel path. FIG. [Figure 13] FIG. 13 is an explanatory diagram of a process for extending an approximation line to the road width. [Figure 14] An example of a cable model created from point cloud N is shown below. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, the embodiments of the present disclosure will be described in detail with reference to the drawings. Note that the present disclosure is not limited to the embodiments shown below. These implementation examples are merely illustrative, and the present disclosure can be implemented in various forms with various modifications and improvements based on the knowledge of those skilled in the art. Note that components with the same reference numerals in this specification and drawings are assumed to indicate the same components.

[0011] An overview of the system according to the present disclosure is shown in Fig. 1. The system according to the present disclosure includes an MMS 20 and an apparatus 10 according to the present disclosure. The apparatus 10 acquires point cloud data measured by the MMS 20 mounted on a vehicle and travel trajectory data, which are the coordinates of the vehicle traveled when the point cloud data was measured, and stores these in association with each other. The point cloud data is data that represents the shape of a structure as a set of points, and each point has three-dimensional coordinates.

[0012] In the present disclosure, an example is shown in which a cable 92 is stretched between utility poles 91-1 and 91-2 and between utility poles 91-2 and 91-3. A vehicle equipped with an MMS 20 runs under the cable 92 stretched between utility poles 91-1 and 91-2. In the present disclosure, an example is shown in which a vehicle runs, but the means of transportation may be any means of transportation capable of running on a road, such as a motorcycle, and the running trajectory may be the running trajectory of any means of transportation. In addition, in the present embodiment, an example is shown in which a point cloud of the cable 92 is extracted, but the point cloud is not limited to the point cloud of the cable 92, which is an overhead cable stretched in the air, and may be any linear object. Hereinafter, when it is not necessary to distinguish between utility poles 91-1 to 91-3, they will be referred to as utility pole 91.

[0013] 2 shows an example of point cloud data measured by the MMS 20. Since there are many more point clouds within a certain distance of a utility pole 91 than there are of a cable 92, it is possible to identify the utility pole 91. This makes it possible to identify the measured point cloud of the utility pole 91 (white circle). Points d1 to d7 indicated by black circles are point clouds other than the utility pole 91.

[0014] FIG. 3 shows an example of a method for extracting linear objects according to the present disclosure. In the present disclosure, a vehicle travel trajectory is used to detect whether or not a point group exists on the travel trajectory (S1). In this embodiment, a point d2 exists on the travel trajectory. In this case, the point d2 is set as a first point group N. FIG. 4 shows a specific example of the first point group N.

[0015] Here, point d2 is a point whose x and y coordinates parallel to the earth's surface among three-dimensional coordinates coincide with the x and y coordinates of the travel path of the vehicle equipped with MMS 20. The x and y coordinates of point d2 may completely coincide with the x and y coordinates of the travel path, but may be within a predetermined range from the x and y coordinates of the travel path.

[0016] In the MMS 20, the cable 92 is measured as a single connected line with multiple points. The intervals between the multiple points vary depending on the vehicle's traveling speed and the distance from the vehicle. Therefore, as shown in FIG. 5, a certain distance D from a point d2 belonging to the first point group N is N It is detected whether or not a point group exists within the fixed distance D. The detected point group is set as the second point group M. N may be any distance that the cable 92 may encounter when measured by the MMS 20.

[0017] Also, since the cable 92 is thin, not many point clouds are measured for the cable 92. Therefore, when the second point cloud M is equal to or less than a predetermined number, it is determined that the first point cloud N is a point cloud of cable candidates. Then, when it is determined that the first point cloud N is a point cloud of cable candidates, each of the second point clouds M is regarded as the first point cloud N. Then, by repeating detection of the second point cloud M for each of the first point clouds N, a point cloud group of cable candidates is created (S2). In this embodiment, a point cloud group of cable candidates of points d1 to d6 shown in FIG. 2 is created.

[0018] Next, line detection is performed on the points included in the point cloud group of the cable candidate, and the point cloud group is divided for each detected line, thereby creating a point cloud group for each cable (S3). For example, as shown in Fig. 6, a line SL1 connecting points d1, d2, and d3, and a line SL2 connecting points d4, d5, and d6 are detected. Then, the point cloud group is divided into a point cloud group (d1, d2, d3) for the line SL1 and a point cloud group (d4, d5, d6) for the line SL2. Here, any algorithm such as Hough transform can be used for the line detection.

[0019] Next, the approximation line obtained by fitting for each point cloud group of the cable is extended, and the point clouds on the extended approximation line are included in the same point cloud group (S4). For example, as shown in FIG. 7, an approximation line FL1 obtained by fitting points d1, d2, and d3 is obtained, and an approximation line FL2 obtained by fitting points d4, d5, and d6 is obtained. Then, a point d7 located on the approximation line FL2 is added to the point cloud group of points d4, d5, and d6. This makes it possible to extract point clouds that are not detected in step S2 because the distance between the points is very far. Therefore, the present disclosure makes it possible to extract the point cloud of the cable 92.

[0020] 8 shows an example of a system configuration according to the present disclosure. The system according to the present disclosure includes an apparatus 10 according to the present disclosure and an MMS 20. The apparatus 10 includes an external data reading unit 11, a setting value storage unit 15, a point cloud determination unit 12, a parameter calculation unit 13, and a calculation result storage unit 14. The apparatus 10 can also be realized by a computer and a program, and the program can be recorded on a recording medium or provided via a network.

[0021] The external data reading unit 11 has a function of reading point cloud data and driving trajectory data from the MMS 20. The set value storage unit 15 has a function of storing parameters used in the processing of the point cloud determination unit 12. The parameters include, for example, a search range value that determines the range of the point cloud data to search for the point d2 in step S2, a distance from the travel path when detecting the point d2 in step S2, a "fixed distance D" used in step S2, N the upper limit of the number of point groups that determines the "predetermined number" used in step S2; and the approximation method used when determining the approximation line in step S4. The point cloud determination unit 12 has functions to detect point clouds on the travel trajectory (S1), create point cloud groups of cable candidates (S2), divide into point cloud groups on a cable basis (S3), generate approximation lines, and detect point clouds on the approximation lines (S4). The parameter calculation unit 13 has a function of calculating the cable height above ground, sag, and span length. The calculation result storage unit 14 has a function of storing the values ​​calculated by the point group determination unit 12 and the parameter calculation unit 13.

[0022] The external data reading unit 11, the set value storage unit 15, the parameter calculation unit 13, and the calculation result storage unit 14 can be realized by known components.

[0023] FIG. 9 shows an example of the process of the device 10. The external data reading unit 11 reads the point cloud data and the driving trajectory data from the MMS 20 (S01), and the point cloud determination unit 12 reads the setting values ​​from the setting value storage unit 15 (S02). In step S1, a detection range is set on the travel locus based on the travel locus information and the search range value (S11), and a point group N existing in the point cloud search range is searched for (S12). In this way, a point group on the travel locus is detected.

[0024] In step S2, the point cloud determination unit 12 executes the following steps S21 to S25. N Point clouds within the inter-point cloud distance are detected and the number of detected point clouds is counted (S21). If the count number is greater than a set value, the point cloud is not determined to be a cable candidate. If the count number is less than the upper limit for the number of point clouds, point clouds N and M are grouped into point cloud group G (S22). For point cloud M as well, the number of point clouds L within the inter-point cloud distance is counted (S23). If the count number is less than the upper limit for the number of point clouds, point cloud L is grouped into point cloud group G (S24). This detection and counting of point clouds within the inter-point cloud distance is repeated to create point cloud group G (S25).

[0025] Depending on the wiring of the cable, it is possible that the point cloud group will be a T-shape. In step S3, the point cloud determination unit 12 detects straight lines (S31) and divides the straight lines into point cloud groups for each cable (S32). In step S4, the point cloud determination unit 12 approximates the point cloud group for each cable with, for example, a linear line, a quadratic curve, or a catenary curve (S41) and extends it (S42). The point clouds present on the extension are also assigned to the point cloud group for each cable (S43) and are used as a cable model (S44). Figure 10 shows an example of a cable model created from the point cloud N in Figure 4. Finally, the parameter calculation unit 13 calculates the cable height above ground, sag, span length, etc. from the cable model (S5). The calculation result storage unit 14 stores the cable model created by the point cloud determination unit 12 and the calculation results of the parameter calculation unit 13.

[0026] Since no utility pole 91 is located on the road on which the vehicle travels, the cable 92 crosses the road. Therefore, the approximation line obtained in step S4 may be extended to the width of the road so as to cross the road. An example of the processing of the point cloud determination unit 12 in this case is shown in Fig. 11. Steps S1 to S3 and S5 are the same as those in Fig. 9, so only step S4 will be described with reference to Figs. 12 and 13.

[0027] An approximation formula is created for each point cloud group in units of cables based on the point cloud coordinates and an approximation method (S41). The coordinates c11 to c16 of the travel path are acquired (S421). The intervals between the coordinates c11 to c16 are arbitrary, and can be set to equal intervals such as every 1 m, for example. The coordinates c21 to c23 of the utility pole 91 are acquired (S422). Here, the coordinates c21 to c23 of the utility pole 91 may be acquired from an equipment DB that stores the geographical position of the utility pole 91, or may be acquired from the coordinates of a utility pole model created using the measured point cloud of the utility pole 91 (the white circle shown in FIG. 2). The coordinates c21 and c22 of the nearest utility poles 91-1 and 91-2 are searched for from the coordinates c11 to c16 of each travel locus acquired in step S421 (S423). A distance D91 between the coordinates c11 to c16 of the travel path and the coordinates c21 and c22 of the utility poles 91-1 and 91-2 is calculated and is set as the road width (S424). As shown in FIG. 13, the approximation line FL1 is extended to the road width D91 (S425). If there is a point group on the approximation line FL1, the point group is made to belong to the point group of points d1, d2, and d3 (S43). In this embodiment, an approximation line FL2 exists on the extended approximation line FL1. In this case, the approximation line FL1 and the approximation line FL2 are connected. The point cloud coordinates of the point cloud group for each cable are stored as cable model information (S44). At this time, the approximation lines FL1 and FL2 are stored in the calculation result storage unit 14 as a cable model connected in a T-shape.

[0028] As described above, the present disclosure uses travel path data, which is the coordinates of the travel of a vehicle equipped with an MMS 20, to detect a point group N of the cable 92 present on the travel path, and calculates a distance D from the point group N. N A point cloud group of the cable candidate is determined based on the number of points, and a point cloud of the cable portion is extracted based on the point cloud included in the point cloud group of the cable candidate.

[0029] In the present disclosure, linear objects are extracted starting from point clouds existing on the travel trajectory, so even if the number of point clouds is sparse, it is possible to extract point clouds of the cable portion and calculate slackness, etc. from the cable model information. Note that, in the specific example of the cable model in FIG. 10, an example is shown in which point cloud N is arranged on the straight line of the cable, but in the present disclosure, as shown in FIG. 14, a point cloud N is arranged on the straight line of the cable, and the present disclosure also allows calculation of slackness, etc. from the cable, starting from point clouds existing on the travel trajectory ... cable N If a point group exists within this range, the effects of the present disclosure can be obtained. [Industrial Applicability]

[0030] The present disclosure can be applied to the information and communications industry. [Explanation of symbols]

[0031] 10: Equipment 11: External data reading section 12: Point cloud determination section 13: Parameter calculation section 14: Calculation result storage section 15: Setting value memory section 20: MMS 91, 91-1, 91-2, 91-3: Electric pole 92: Cable

Claims

1. Acquire point data representing three-dimensional coordinates and a movement trajectory when measuring the point data, determining utility pole points based on a density of points to exclude utility pole points from the point data; Detecting a specific point located immediately above the movement trajectory from the excluded point data; detecting points within a certain distance from the specific point as a candidate group; Counting the number of points included in the candidate group; When the count number is equal to or less than a predetermined number, the specific point is determined to be a linear object candidate; Each point included in the candidate group of the specific point determined to be a linear object candidate is regarded as the specific point, and the detection of the candidate group and the counting are repeated for each of the points regarded as the specific point, thereby extracting points of the linear object candidate; Detecting straight lines from the extracted points of the linear object candidates, grouping each detected straight line, obtaining an approximation line from the grouped points, and creating a three-dimensional model of the linear object using the points obtained by including points located on the approximation line in the same group. Device.

2. A linear line, a quadratic curve, or a catenary curve is used as the approximation line.

2. The apparatus of claim 1.

3. The linear object is an overhead cable crossing a road, Extending the approximation line so as to cross the road; 3. The apparatus of claim 2.

4. Acquire point data representing three-dimensional coordinates and a movement trajectory when measuring the point data, determining utility pole points based on a density of points to exclude utility pole points from the point data; Detecting a specific point located immediately above the movement trajectory from the excluded point data; detecting points within a certain distance from the specific point as a candidate group; Counting the number of points included in the candidate group; When the count number is equal to or less than a predetermined number, the specific point is determined to be a linear object candidate; Each point included in the candidate group of the specific point determined to be a linear object candidate is regarded as the specific point, and the detection of the candidate group and the counting are repeated for each of the points regarded as the specific point, thereby extracting points of the linear object candidate; Detecting straight lines from the extracted points of the linear object candidates, grouping each detected straight line, obtaining an approximation line from the grouped points, and creating a three-dimensional model of the linear object using the points obtained by including points located on the approximation line in the same group. method.

5. A program for causing a computer to realize each of the functional units of the device according to any one of claims 1 to 3.

Images