JP2023034604A - Road plane point group data correction method and device thereof - Google Patents

Abstract

To provide a method of correcting road plane point group data generated from point group data obtained by a mobile mapping system, and a device thereof.SOLUTION: A road plane point group data correction device 1 comprises: a point group data acquisition unit which acquires point group data corresponding to a road; a general road plane point group data generation unit which processes the point group data to obtain, from the point group data, first road plane point group data which corresponds to a road surface on which vehicles travel and is obtained by normalizing it to a plane; a virtual point group data extraction unit which extracts virtual point group data corresponding to a virtual road from the point group data; a virtual road plane point group data generation unit which processes the virtual point group data corresponding to the virtual road to obtain, from the virtual point group data, second road plane point group data which corresponds to a road surface of the virtual road and is obtained by normalizing it to a plane; and a road plane point group data correction unit which corrects the first road plane point group data with the second road plane point group data.SELECTED DRAWING: Figure 3

Description

The present invention relates to a method and apparatus for correcting road plane point cloud data generated from point cloud data obtained by a mobile mapping system.

Map data is created by processing point cloud data obtained by a mobile mapping system (sometimes abbreviated as "MMS" in this specification) consisting of a 3D laser measuring machine and a digital camera (patent References 1-4).
The point data constituting the point cloud data are given a coordinate y in the traveling direction of the photographing vehicle, a coordinate x in a direction perpendicular to the traveling direction, and a coordinate z in the height direction.
Data specifying road planes in map data (referred to as "road plane point group data" in this specification) is generated by computer processing that executes the following algorithm, for example.

Extraction of ground surface point cloud data For all point cloud data existing at the same coordinate yn in the traveling direction of the photographing vehicle (including those on the road surface and those on the non-road surface), the extraction center is (yn, xn) , points (primary extraction point cloud data) existing within a predetermined primary extraction width ±A in the transverse direction from the extraction center are extracted. Let the median value zn of the heights of the primary extraction point cloud data extracted for each extraction center (yn, xn) be the reference height zn at the extraction center (yn, xn). This reference height zn is specified for all the point group data (both on the road surface and on the non-road surface) present at the same coordinate yn.
Next, a predetermined secondary extraction height width ±B is given around the reference height zn specified for each extraction center (yn, xn), and the Extract point cloud data. The point cloud data (ground surface point cloud data) extracted in this manner is indicated by symbol P in FIG. This ground surface point cloud data covers the entire ground surface photographed by MMS.

Extraction of road plane point cloud data Extract road plane point cloud data from the ground surface point cloud data. Although the cross-sectional shape of the actual road surface is curved, it is assumed that the cross-sectional shape is linear, that is, the road surface is flat, and the road plane point cloud data is extracted. Note that the road surface refers to an area where the vehicle is scheduled to travel.
Link data and road width data of road map data provided in advance are used to select and extract data corresponding to the road surface from the ground surface point group data. Here, in the example of FIG. 1, the point cloud data included in the extracted cross-sectional area indicated by symbol E corresponds to the road surface. The height of the extracted cross-sectional area E is in the range of the secondary extracted height width 2B, and the width of the extracted cross-sectional area E is the width of the road surface obtained from the road map data. It can be said that such extraction cross-sectional areas E are connected along the link to form a flat extraction area.

Since the reference height zn specified from the primary extracted point cloud data varies in the road width direction, the reference height zn in the road width direction is normalized in order to assume that the road surface is planar. For example, the least squares method or the like is used to place the reference height zn at each extraction center (yn, xn) on a straight line. As a result, the extracted cross-sectional area E becomes rectangular.
It is assumed that the point cloud data contained in the extracted cross-sectional area E corresponds to the road surface, and the extracted cross-sectional area E is used as a rectangular filter to normalize the road surface to a plane. The point cloud data extracted in this manner is referred to as road plane point cloud data in this specification.

JP 2019-027973 A JP 2018-004477 A JP 2014-115904 A Japanese patent publication WO2010/024212

The road plane point cloud data automatically generated by the computer is manually corrected by the operator and reflected in the map data.
When the road plane point cloud data was automatically created as described above, the boundary line between the road surface on which the vehicle should originally pass and the non-road surface (road shoulder, median strip, etc.) that is continuous with the edge of the road surface was clearly demarcated. There was nothing. For example, the edge area of the road surface is missing from the road surface data group.
It takes a lot of time and effort to manually correct such omissions. Therefore, it is required to reduce such omissions.

The reason why the lack occurs is considered as follows.
On an actual road, non-road surfaces (road shoulders and median strips) may incline more in the vertical direction than the road surface. Here, when the primary extraction width (±A) in the road crossing direction is large, the slope influences the specification of the reference height zn.
For example, when the extraction center is near the non-road surface on the road surface, if the non-road surface has an inclination, the area with this inclination is included in the primary extraction width ±A. As a result, the slope region has a great influence on specifying the reference height zn of the extraction center on the road surface (see region P1 in FIG. 1). However, the influence of the inclination on the extracted cross-sectional area E is small. This is because the inclination of the extracted cross-sectional area E is specified by normalizing the reference height zn of the entire road surface, and therefore the influence of other parts of the road surface (the wide part on the central side) is greater than that of the area P1. be. As a result, the extracted cross-sectional area E deviates from the P1 area, causing lack of point cloud data.

The present invention has been made to solve the above problems, and the first aspect thereof is defined as follows. Namely
obtaining point cloud data corresponding to the specified road;
generating ground surface point cloud data corresponding to the road from the point cloud data;
The road information of the road map data is referred to, the ground surface point cloud data is processed by a predetermined algorithm, the road surface on which the vehicle travels is corresponded from the ground surface point cloud data, and the road surface is normalized to a plane. obtaining first road plane point cloud data;
With reference to the road information of the road map data and the information defining one side edge of the virtual road, the ground surface point cloud data is processed by a predetermined algorithm to obtain a virtual road from the ground surface point cloud data. a step of obtaining second road plane point cloud data corresponding to the road surface of and normalized to a plane;
a correction step of correcting the first road plane point cloud data with the second road plane point cloud data;
A method for correcting road plane point cloud data.

According to the correction method of the road plane point cloud data of the first aspect defined in this way, a virtual road is set, and the road plane point cloud data is corrected using a predetermined algorithm for this virtual road. create. This virtual road has one side edge within the travel area of the road surface and the other side edge coincides with the side edge of the road surface. That is, it is narrowed on the center side of the road surface. As a result, in normalizing the reference height zn of the extracted cross-sectional area E, that is, when specifying the inclination angle of the extracted cross-sectional area E, the influence of the center side of the road surface is reduced.

FIG. 2 shows extracted cross-sectional areas E1 and E2 set for a virtual road.
The influence of the inclination of the non-road surface appears greatly in the extracted cross-sectional area E1 of FIG. This is because the influence on the central side of the road surface is smaller than in the example of FIG.
As a result, the extracted cross-sectional area E1 covers the point cloud data of the area P1 that was not covered through the extracted cross-sectional area E.

By supplementing the point cloud data of the region lacking the first road plane point cloud data extracted through the extraction cross-sectional region E with the second road plane point cloud data extracted through the extraction cross-sectional region E1, the road plane point cloud is obtained. The data corresponds to the actual road surface. Therefore, in the road surface image displayed based on the road plane point cloud data, the boundary between the road surface and the non-road surface can be clearly distinguished. Also, repair work by the operator is reduced.

In this way, correcting the point cloud data covering the entire road surface with the point cloud data covering a part of the road surface including the side edges of the road surface is effective when the normalization process is performed to make a plane. The non-road surface that follows the side edge of the road surface may incline more than the road surface, and it is impossible to distinguish between the road surface and the non-road surface in the point cloud data (before processing) captured by the MMS imaging device. Therefore, while the inclination of the non-road surface affects the point cloud data processing method, the non-road surface is narrower than the road surface, so the influence of the inclination of the non-road surface on the normalization process for flattening the road surface is small.

In setting the virtual road, the other side edge of the virtual road is matched with the side edge of the general road (specified in the road map data), and one side is within the driving area of the general road surface. be. One side of such a virtual road surface can be arbitrarily set within the general road surface (that is, the running surface), and can be matched with the movement locus of the MMS imaging device. This is because it is easy to record the position at the time of MMS imaging, which facilitates data handling.

An apparatus for carrying out the first phase is defined as follows.
a point cloud data acquisition unit for acquiring point cloud data corresponding to a specified road;
a ground surface point cloud data generation unit that generates ground surface point cloud data corresponding to the road from the point cloud data;
The road information of the road map data is referred to, the ground surface point cloud data is processed by a predetermined algorithm, the road surface on which the vehicle travels is corresponded from the ground surface point cloud data, and the road surface is normalized to a plane. a general road plane point cloud data generator for obtaining the first road plane point cloud data;
With reference to the road information of the road map data and the information defining one side edge of the virtual road, the ground surface point cloud data is processed by a predetermined algorithm to obtain a virtual road from the ground surface point cloud data. a virtual road plane point cloud data generation unit that obtains second road plane point cloud data corresponding to the road surface of and normalized to a plane;
a road plane point cloud data correction unit that corrects the first road plane point cloud data with the second road plane point cloud data;
A correction device for road plane point cloud data.

A computer program for operating the computer system for this correction device is defined as follows.
Correction of road plane point cloud data comprising a point cloud data acquisition device, a ground surface point cloud data generation unit, a general road plane point cloud data generation unit, a virtual road plane point cloud data generation unit, and a road plane point cloud data correction unit A computer program applied to a computer device constituting a device,
causing the point cloud data acquisition device to acquire point cloud data corresponding to the specified road;
causing the ground surface point cloud data generation unit to generate ground surface point cloud data corresponding to the road from the point cloud data;
The general road plane point cloud data generation unit refers to the road information of the road map data, processes the ground surface point cloud data with a predetermined algorithm, and converts the ground surface point cloud data into the road surface on which the vehicle travels. obtaining the first road plane point cloud data corresponding to and normalized to the plane;
The virtual road plane point cloud data generation unit refers to the road information of the road map data and the information defining one side edge of the virtual road, and processes the ground surface point cloud data with a predetermined algorithm, obtaining second road plane point cloud data corresponding to the road surface of a virtual road and normalized to a plane from the ground surface point cloud data;
A computer program for causing the road plane point cloud data correction unit to correct the first road plane point cloud data with the second road plane point cloud data.

A second aspect of the invention is defined as follows.
obtaining point cloud data corresponding to the road;
a step of processing the point cloud data with a predetermined algorithm to obtain, from the point cloud data, first road plane point cloud data corresponding to the road surface on which the vehicle travels and normalized to a plane;
an extracting step of extracting virtual point cloud data corresponding to a virtual road from the point cloud data, wherein one side edge of the virtual road is within a travel area of the road surface and the other side edge is coincides with the side edge of the road surface, a virtual point cloud data extraction step;
A second road obtained by processing the virtual point cloud data corresponding to the virtual road with the predetermined algorithm to correspond to the road surface of the virtual road from the virtual point cloud data and normalizing it to a plane. obtaining planar point cloud data;
a correction step of correcting the first road plane point cloud data with the second road plane point cloud data;
A method for correcting road plane point cloud data.

A correction device for executing the correction method of the second aspect is defined as follows.
a point cloud data acquisition device for acquiring point cloud data corresponding to a road;
The point cloud data is processed by a predetermined algorithm to obtain general road plane point cloud data corresponding to the road surface on which the vehicle travels and normalized to a plane from the point cloud data. a data generator;
a virtual point cloud data extraction unit for extracting virtual point cloud data corresponding to a virtual road from the point cloud data, wherein one side edge of the virtual road is within a travel area of the road surface; a virtual point cloud data extraction unit, the other side edge of which coincides with the side edge of the road surface;
A second road obtained by processing the virtual point cloud data corresponding to the virtual road with the predetermined algorithm to correspond to the road surface of the virtual road from the virtual point cloud data and normalizing it to a plane. a second road plane point cloud data generator for obtaining plane point cloud data;
a road plane point cloud data correction unit that corrects the first road plane point cloud data with the second road plane point cloud data;
A correction device for road plane point cloud data.

A computer program for executing the computer device of this correction device is defined as follows.
Road plane point cloud data acquisition device comprising a point cloud data acquisition device, a first road plane point cloud data generation unit, a virtual point cloud data extraction unit, a second road plane point cloud data generation unit, and a road plane point cloud data correction unit A computer program applied to a computer device constituting a correction device,
causing the point cloud data acquisition device to acquire point cloud data corresponding to the road;
The first road plane point cloud data generation unit processes the point cloud data with a predetermined algorithm, corresponds to the road surface on which the vehicle travels from the point cloud data, and normalizes it to a plane. Obtain planar point cloud data,
causing the virtual point cloud data extracting unit to extract virtual point cloud data corresponding to a virtual road from the point cloud data, wherein one side edge of the virtual road is within a travel area of the road surface; The side edge of the coincides with the side edge of the road surface,
The second road plane point cloud data generation unit processes the virtual point cloud data corresponding to the virtual road with the predetermined algorithm to generate a road surface of the virtual road from the virtual point cloud data. , and obtain the second road plane point cloud data normalized to the plane,
A computer program for causing the road plane point cloud data correction unit to correct the first road plane point cloud data with the second road plane point cloud data.

FIG. 1 is a schematic diagram for explaining a conventional method for generating road plane point cloud data. FIG. 2 is a schematic diagram for explaining a method of generating road plane point cloud data of a virtual road according to the present invention. FIG. 3 is a block diagram showing an apparatus for correcting road plane point cloud data according to an embodiment of the present invention. FIG. 4 is a flowchart for explaining a method of correcting road plane point cloud data using the device of FIG. FIG. 5 is a flow chart showing details of step 3 in FIG. FIG. 6 is a flow chart showing details of step 5 in FIG. FIG. 7 is a functional block of a computer device that operates the road plane point cloud data correction device of the embodiment.

The block diagram of FIG. 3 shows the configuration of the road plane point cloud data correction device 1 of the embodiment.
The correction device 1 includes a database unit 10, a data storage unit 20, a ground surface point cloud data generation unit 25, a road plane point cloud data generation unit 30, a road plane point cloud data storage unit 40, and a road plane point cloud data correction unit 50. roughly comprising:
The database unit 10 includes a point cloud database 11 that stores all point cloud data captured by the MMS imaging device, and an MMS trajectory database 13 that stores the trajectory (coordinates) of the MMS imaging device.
The MMS imaging device generally consists of a three-dimensional laser measuring device and a digital camera mounted on a vehicle. Here, the trajectory of the MMS imaging device is specified from the travel data of the vehicle. The drone can also be equipped with an MMS imager.

The operator refers to link data of roads stored in the road map data storage unit 70 using the link specifying unit 60 and specifies link data for which road plane point cloud data is to be generated. At this time, when the y-coordinates of the link data are y1 to yt, all x-coordinate and z-coordinate data are specified for each y-coordinate. In other words, not only the point cloud data corresponding to the road surface but also the point cloud data corresponding to the non-road surface are designated together.
Point cloud data corresponding to the designated link data is read from the point cloud database 11 and stored in the point cloud data storage unit 21 . Similarly, the trajectory data of the MMS photographing device corresponding to the designated link data is read out from the MMS trajectory database 13 and stored in the MMS trajectory storage unit 23 .

The ground surface point cloud data generation unit 25 generates ground surface point cloud data from the point cloud data stored in the point cloud data storage unit 21 .
The general road plane point cloud data generation unit 31 forms an extracted cross-sectional area E by referring to the coordinates of the link data of the road map data and the road width data (see FIG. 1). From the extracted ground surface point cloud data, the point cloud data included in the extracted cross-sectional area E are extracted and used as the first road plane point cloud data.
The first road plane point cloud data thus generated is stored in the first road plane point cloud data storage unit 41 .

The virtual road plane point cloud data generation unit 33 refers to the coordinates of the link data of the road map data and the trajectory data of the MMS imaging device stored in the MMS trajectory storage unit 23, and forms an extracted cross-sectional area E1 (Fig. 2 reference). , from the ground surface point cloud data generated by the ground surface point cloud data generation unit 25, the point cloud data included in the extracted cross-sectional area E1 is extracted, and is used as the second road plane point cloud data.
The second road plane point cloud data generated in this manner is stored in the second road plane point cloud data storage unit 43 .
In this example, one side edge of the virtual road is automatically determined based on the trajectory of the MMS imaging device, but the operator can arbitrarily determine one side edge of the virtual road. It should be noted that this one side edge is preferably located substantially in the center of the general road.
In the above, a virtual road is defined in advance by referring to the coordinates of the link data of the road map data and the trajectory data of the MMS camera stored in the MMS trajectory storage unit 23, and stored in the point cloud database 11. Point cloud data corresponding to the virtual road is extracted from the point cloud data, and the extracted point cloud data is processed by the ground surface point cloud data generation unit 25 and the general road plane point cloud data generation unit 31. By executing similar processing, the second road plane point cloud data can also be generated.

The road plane point cloud data correction unit 50 corrects the first road plane point cloud data stored in the first road plane point cloud data storage unit 41 and the second road plane point cloud data stored in the second road plane point cloud data storage unit 43 . The second road plane point cloud data is compared, and data missing in the first road plane point cloud data is supplemented with the second road plane point cloud data.
In this example, when the second road plane point cloud data exists at the coordinates where the first road plane point group data is missing, the former data is supplemented by inserting the latter data into the former data.
In addition, the data of the coordinates around the missing coordinates in the first road plane point cloud data can also be replaced with the second road plane point cloud data.
Corresponding the second road plane point cloud data to the first road plane point cloud data in this manner is referred to as correction in this specification.

Next, the operation of the road plane point cloud data correction device 1 shown in FIG. 3 will be described with reference to the flowcharts shown in FIGS.
In step 1, the link designating unit 60 reads from the point cloud database 11 the point cloud data corresponding to the link data (coordinates y1 to yt) of the road designated by the operator or according to a predetermined rule. Store in the storage unit 21 . Similarly, the link designation unit 60 reads out the movement trajectory of the MMS imaging device corresponding to the link data of the designated road from the MMS trajectory database 13 and stores it in the MMS trajectory storage unit 23 .

In step 2, ground surface point cloud data is generated.
When the extraction center is (y1, xn) for all the point cloud data existing at the same coordinate y1 in the traveling direction of the MMS imaging device (including both road surface and non-road surface), the extraction center is points (primary extraction point cloud data) existing within a predetermined primary extraction width ±A (for example, 1.0 m) in the transverse direction. Let the median value zn of the heights of the primary extraction point cloud data extracted for each extraction center (y1, xn) be the reference height zn at the extraction center (y1, xn). This reference height zn is specified for all the point group data (both on the road surface and on the non-road surface) present at the same coordinate y1.
Next, a predetermined secondary extraction height width ±B (for example, 0.5 m) is given around the reference height zn specified for each extraction center (y1, xn), and the coordinates (y1, xn, zn ±B) to extract point cloud data. The ground surface point cloud data extracted in this manner is indicated by symbol P in FIG.

The general road plane point cloud data generation unit 31 generates road plane point cloud data at the coordinate y1, which is the starting point of the link data. This road plane point cloud data is stored in the first road plane point cloud data storage unit 41 as first road plane point cloud data.
More specifically, in steps 33 and 35, road plane point cloud data is extracted from the ground surface point cloud data generated in step 2. FIG.
The general road plane point group data generator 31 refers to the coordinates of the link data of the road map data stored in the road map data storage unit 70 and the road width data to form an extracted cross-sectional area E (see FIG. 1). The height (length in the direction of the vertical axis) of the extracted cross-sectional area E is the secondary extraction height width 2B, and its width (length in the direction of the horizontal axis) is specified from the road width data (step 33). .
After that, in step 35, the ground surface point cloud data included in the extracted cross-sectional area E are extracted.
Conventional road plane point cloud data is generated based on the point cloud data thus extracted.

In step 5, the virtual road plane point cloud data generating unit 31 generates virtual road plane point cloud data at the coordinate y1, which is the starting point of the link data. This data is obtained by running an existing algorithm (same as used in step 3) on the established virtual road.
This virtual road plane point cloud data is stored in the second road plane point cloud data storage unit 43 as second road plane point cloud data.
Details of step 5 are shown in FIG. The ground surface point cloud data extracted in step 2 is indicated by symbol P in FIG.

In steps 53 and 55, virtual plane point cloud data is extracted from the ground surface point cloud data.
More specifically, the virtual road plane point cloud data generation unit 33 refers to the road map data stored in the road map data storage unit 70 and the trajectory data of the MMS imaging device stored in the MMS trajectory storage unit 23. A cross-section extraction region E1 is formed (see FIG. 2). The height (length in the direction of the vertical axis) of the cross-section extraction region E1 is the secondary extraction height width 2B, and the width (length in the direction of the horizontal axis) is the road width data of the road map data and the MMS camera. and the trajectory of (step 53). In other words, one side edge of the virtual road within the travel area of the road surface is defined by the trajectory of the MMS imager. The other side edge of the virtual road is matched with the side edge of the road specified from the link coordinates and road width data.
After that, in step 55, the ground surface point cloud data included in the extracted cross-sectional area E1 are extracted. The extracted point cloud data is stored in the second road plane point cloud data storage unit 43 as second road plane point cloud data.

In the above, the road map data stored in the road map data storage unit 70 and the trajectory data of the MMS photographing device stored in the MMS trajectory storage unit 23 are referenced to set virtual roads in advance. By regarding this virtual road as a general road, the surface point cloud data corresponding to the virtual road is read out from the point cloud database 11, and the same processing as in step 3 is performed to obtain the second road plane Point cloud data can also be obtained.

In step 7, the road plane point cloud data correction unit 50 stores the first road plane point cloud data stored in the first road plane point cloud data storage unit 41 and the second road plane point cloud data storage unit 43. The data missing in the former are supplemented with the latter data by comparing with the second road plane point cloud data.
This operation is repeated up to the coordinate yt of the link data (steps 9 and 10).
Correction of the point cloud data may be performed after executing steps 3 and 5 for the coordinates y1 to yt.
The same processing as described above can be performed for the cross-section extraction region E2 in FIG.

FIG. 7 shows the hardware configuration of the correction device 1. As shown in FIG.
A computing unit 300 includes a CPU 301, a ROM 303 and a RAM 305, and controls the entire system. At the same time, it functions as a ground surface point cloud data generation unit 25 , a road plane point cloud data generation unit 30 , a road plane point cloud data correction unit 50 and a link designation unit 60 . A ROM 303 is a non-volatile memory that stores control programs and the like for controlling the arithmetic unit 300 . The RAM 305 readablely stores various setting values preset by the user through an input device 330 such as a keyboard, and provides a working area for the CPU 301 . The control program for controlling the arithmetic unit 300 may be stored in the RAM 305 or the first and second storage devices 340 and 350 instead of the ROM 303 . Various correlation data are sent to the navigation device via the output device 320 . Conditions for acquiring vehicle travel data are input via the input device 330 .

The first storage device 340 functions as the data storage unit 20 and the road plane point cloud data storage unit 40 .
The second storage device 350 can function as the database unit 10 and the road map data storage unit 70, but the storage of these data can also be performed by an external server.
Preferably, the first and second storage devices use a partial area of the memory device of the server system, such as hard memory or flash memory.

A partial area of the RAM of the calculation unit can be used as a so-called buffer memory as a storage unit for temporarily stored data (ground surface point cloud data generated by the road plane point cloud data generation unit, etc.).
Each device constituting the computer is connected by a system bus 370 .

The present invention is by no means limited to the description of the above embodiments and examples of the invention. Various modifications are also included in the present invention within the scope of those skilled in the art without departing from the description of the claims.

1 road plane point cloud data correction device 10 database unit 20 data storage unit 25 ground surface point cloud data generation unit 30 road plane point cloud data generation unit 40 road plane point cloud data storage unit 50 road plane point cloud data correction unit
P Ground surface point cloud data E, E1, E2 Extracted cross-sectional area

Claims (5)

obtaining point cloud data corresponding to the specified road;
extracting ground surface point cloud data corresponding to the road from the point cloud data;
The road information of the road map data is referred to, the ground surface point cloud data is processed by a predetermined algorithm, the road surface on which the vehicle travels is corresponded from the ground surface point cloud data, and the road surface is normalized to a plane. obtaining first road plane point cloud data;
With reference to the road information of the road map data and the information defining one side edge of the virtual road, the ground surface point cloud data is processed by a predetermined algorithm to obtain a virtual road from the ground surface point cloud data. a step of obtaining second road plane point cloud data corresponding to the road surface of and normalized to a plane;
a correction step of correcting the first road plane point cloud data with the second road plane point cloud data;
A method for correcting road plane point cloud data. 2. The correction method according to claim 1, wherein said correcting step fills voids in said first road plane point cloud data with said second road plane point cloud data. 3. The correction method according to claim 1 or 2, wherein one side edge of the virtual road coincides with the movement trajectory of the MMS imaging device. a point cloud data acquisition unit for acquiring point cloud data corresponding to a specified road;
a ground surface point cloud data generation unit that generates ground surface point cloud data corresponding to the road from the point cloud data;
The road information of the road map data is referred to, the ground surface point cloud data is processed by a predetermined algorithm, the road surface on which the vehicle travels is corresponded from the ground surface point cloud data, and the road surface is normalized to a plane. a general road plane point cloud data generator for obtaining the first road plane point cloud data;
With reference to the road information of the road map data and the information defining one side edge of the virtual road, the ground surface point cloud data is processed by a predetermined algorithm to obtain a virtual road from the ground surface point cloud data. a virtual road plane point cloud data generation unit that obtains second road plane point cloud data corresponding to the road surface of and normalized to a plane;
a road plane point cloud data correction unit that corrects the first road plane point cloud data with the second road plane point cloud data;
A correction device for road plane point cloud data. Correction of road plane point cloud data comprising a point cloud data acquisition device, a ground surface point cloud data generation unit, a general road plane point cloud data generation unit, a virtual road plane point cloud data generation unit, and a road plane point cloud data correction unit A computer program applied to a computer device constituting a device,
causing the point cloud data acquisition device to acquire point cloud data corresponding to the specified road;
causing the ground surface point cloud data generation unit to generate ground surface point cloud data corresponding to the road from the point cloud data;
The general road plane point cloud data generation unit refers to the road information of the road map data, processes the ground surface point cloud data with a predetermined algorithm, and converts the ground surface point cloud data into the road surface on which the vehicle travels. obtaining the first road plane point cloud data corresponding to and normalized to the plane;
The virtual road plane point cloud data generation unit refers to the road information of the road map data and the information defining one side edge of the virtual road, and processes the ground surface point cloud data with a predetermined algorithm, obtaining second road plane point cloud data corresponding to the road surface of a virtual road and normalized to a plane from the ground surface point cloud data;
A computer program for causing the road plane point cloud data correction unit to correct the first road plane point cloud data with the second road plane point cloud data.

Images