JP4674051B2 - Method for generating 3D map data - Google Patents

Description

  The present invention relates to a technique for generating three-dimensional electronic map data using at least a part of two-dimensional map data, photograph data obtained by photographing a terrain from a high place, and road network data.

  In recent years, the use of map data digitized so as to be usable by computers (hereinafter referred to as “electronic map data”) has become widespread. The electronic map data is used for displaying a map on a so-called personal computer, a vehicle-mounted navigation system, providing a map via the Internet, and creating a map as a printed matter. In navigation systems, three-dimensional display is being used so that the driver can intuitively determine the course. In the three-dimensional display, in addition to three-dimensional data such as buildings and houses, three-dimensional data that is realistic is required for topography and elevated roads. Various techniques have been proposed for generating three-dimensional map data for enabling three-dimensional display with a light load. For example, Patent Document 1 discloses a method of generating coastline polygon data by combining altitude data prepared with a predetermined mesh and topographic data.

JP 2002-306073 A

  However, the prior art alone is insufficient for the efficient generation of 3D map data, and further, it is necessary to improve efficiency in many ways. For example, the prior art cannot be applied to the generation of terrain data having a vertical cut-off such as a pier at a port. An object of the present invention is to further improve the efficiency of generating three-dimensional map data.

  The present invention can be configured as a generating device that generates three-dimensional map data. Below, the 1st-4th production | generation apparatus is illustrated. The first generation device performs modeling of terrain (hereinafter referred to as “digging process”) in which a vertically cut portion such as a pier at a port exists. A 2nd production | generation apparatus specifies intersection forms, such as elevated, embankment, and cut, about the traffic road which cross | intersects three-dimensionally. In this specification, a traffic road is a general term for passages used for traffic such as roads, sidewalks, and railways. The third generation device specifies the position of the tunnel. The fourth generation device determines whether the elevated traffic road is configured such that the vertical lines are configured in parallel or overlapped in the vertical direction (hereinafter, such a structure is referred to as a vertical elevated road). These may be configured as four types of independent generation apparatuses, or may be combined and integrated as appropriate in one generation apparatus.

  Each generation device uses any of altitude data, map data, traffic network data, and photographic data. The altitude data is an altitude value given to a mesh defined by dividing the ground surface into predetermined areas. The map data means two-dimensional map data including polygon data representing terrain and its attribute data. Photo data is electronic data of a photograph obtained by photographing an area corresponding to map data, such as an aerial photograph, from a high place. The traffic route network data is data used for route search, and is data in which connections of traffic routes are represented by links and nodes. Such data may be stored in the generation apparatus as a database inside, or the generation apparatus may refer to data stored in a server, a recording medium, or the like.

  The 1st production | generation apparatus implement | achieves a digging process in the following procedures. The generation device selects a target polygon to be a target of the mining process from the polygons in the map data. Then, after assigning altitude data of the mesh outside the target polygon located in the vicinity to each mesh including the boundary line of the target polygon, three-dimensional data of the ground surface is generated based on the mesh and the altitude data. . As a result, the elevation data of each mesh including the boundary line of the target polygon is forcibly matched to the neighboring mesh. Thereafter, the altitude value of the area corresponding to the target polygon is corrected to the predetermined value set regardless of the altitude data for the three-dimensional data on the ground surface. By doing so, the elevation of the target polygon region changes discontinuously from the other regions, so that a vertical plane appears on the boundary line of the target polygon. Since the mesh including the boundary of the target polygon is given a nearby elevation, even if the shape of the target polygon and the mesh shape are deviated, it is possible to appropriately form a vertical plane along the boundary of the target polygon. it can.

  The target polygon can be, for example, a polygon representing the sea surface. In this case, the altitude given to the target polygon can be a preset fixed value, for example, a value of 0. By doing so, it is possible to generate a three-dimensional model such as a harbor pier or a cliff. In addition, the target polygon may be a polygon that represents a laid track in a railway station, and the average value of the altitude of a mesh including the station may be used as the altitude. By doing so, it is possible to appropriately generate a three-dimensional model of the station platform.

  The target polygon may be a polygon representing a river. Rivers generally have high altitudes in the inland areas, and the altitudes decrease as they approach the coast. Therefore, it is possible to adopt a method in which the altitude for the river is given as a predetermined relative value with respect to the altitude near the target polygon. The method of setting the altitude of the target polygon with the relative value is not limited to rivers and can be applied to various target polygons such as lakes and ponds.

  Rivers often have bridges. If there is no mesh included in the bridge, the elevation of the bridge may be unknown. In such a case, the bridge may be given an elevation relative to the ground part on both sides of the river. If there are riverbanks and embankments on both sides of the river, it is desirable to give the bridge an elevation corresponding to the embankments instead of the riverbanks. In order to realize such processing, the elevation distribution along the bridge may be obtained, and the elevation of the bridge may be set based on the maximum value.

  In some polygons adjacent to the target polygon, there may be a polygon whose elevation should change continuously from the target polygon (hereinafter referred to as a connected polygon). Examples include polygons on the sea surface for river polygons, and polygons on railway lines outside the station, with respect to polygons on the construction lines inside the station. In the first generation device, as a result of the digging process, discontinuity may occur in the altitude between the target polygon and these connected polygons. In such a case, a connected polygon for ensuring continuity of elevation may be generated between the target polygon and the connected polygon. The connected polygon may be generated by dividing or deforming a part of the target polygon or the connected polygon, or may be newly generated. Among the vertices that define the connected polygons, the vertices located on the boundary of the target polygon or the connected polygon are given the elevations of these polygons. The elevations of the vertices existing outside the boundary between the two may be set so as to interpolate the elevations of the target polygon and the connected polygon. This interpolation may be linear, but it is preferable to use a Bezier curve or a spline curve in order to smoothly connect the target polygon and the connected polygon.

  A 2nd production | generation apparatus specifies intersection forms, such as elevated, embankment, and cut, about the traffic road which cross | intersects three-dimensionally. The second generation device refers to map data and photograph data. Then, based on the map data, a plurality of traffic paths that intersect three-dimensionally are extracted as processing targets. For example, in the map data, when the boundary line of the first traffic path is divided by the boundary line of the second traffic path, the second traffic path passes above the first traffic path. I understand that. However, in this state, the first traffic route is in the basement, that is, it is cut, or is the second traffic route placed higher than the ground surface by overpass or embankment? Cannot be specified. The second generation device, for the processing target, next specifies the positional relationship between the traffic road and at least one of the shadow of the traffic road to be processed and the shadow formed on the traffic road, based on the photograph data, and this positional relationship Based on the above, the intersection type of the traffic road is specified. Since cuts, overpasses, and fills differ in how shadows are created, the form of intersection can be specified by referring to the shadows in this way.

  For example, it is possible to determine that a traffic route in which its own shadow protrudes to the outside is elevated or embankment. For traffic roads determined to be elevated or embankment, map data may be referenced to determine whether the road corresponds to an elevated or embankment based on the distance between the building adjacent to the traffic road and the traffic road. In the case of embankment, since there is no space below the traffic road, the distance between the building and the traffic road inevitably increases, whereas in the case of an elevated structure, this distance is often relatively small. In addition to such determination, a method of determining that the shadow is filled when the shadow is light may be adopted.

  Regarding cuts, it is possible to adopt a method of determining cuts for traffic paths in which the shadows of the processing objects are within. In addition, it is also possible to determine that it is cut by the shadow of the traffic path passing above itself falling within itself.

  The third generation device specifies the position of the tunnel. The third generation device refers to the map data and the traffic route network data. And about the traffic route corresponding to the link of traffic route network data, it identifies that the tunnel is provided in the part represented with the broken line in map data. In the map data, the presence or absence of a line representing the wellhead may be further referred to.

  The fourth generation device identifies the vertical elevated road using the map data and the photograph data. Then, by comparing the two, traffic roads having different numbers of upper and lower lines are identified as vertical elevated roads in which the upper and lower lines are provided on different planes. In contrast to vertical elevated roads, the map data is shown with the vertical lines shifted to the left and right so that the vertical lines can be discriminated, whereas the photo data contains only one of the vertical lines. Therefore, it is possible to determine that the traffic roads having different numbers of upper and lower lines between the map data and the photograph data are vertical elevated roads. As a method for simplifying the above determination, for example, a difference in the number of upper and lower lines may be determined based on a difference in traffic width between map data and photograph data.

  The present invention may be configured as a generation method for generating three-dimensional map data by a computer in addition to the generation device described above. In addition, a computer program for causing a computer to generate three-dimensional map data, and a recording medium on which the computer program is recorded so as to be readable by a computer can also be configured. Recording media include flexible disks, CD-ROMs, magneto-optical disks, IC cards, ROM cartridges, punch cards, printed matter printed with codes such as bar codes, computer internal storage devices (memory such as RAM and ROM), and Various media that can be read by a computer, such as an external storage device, can be used.

Embodiments of the present invention will be described in the following order.
A. Device configuration:
B. Ground polygon generation processing:
C. Mining treatment:
D. Three-dimensional intersection processing:
E. Elevated form specific processing
F. Tunnel processing:
G. Finishing process:

A. Device configuration:
FIG. 1 is an explanatory diagram illustrating a schematic configuration of a three-dimensional map data generation apparatus according to an embodiment. The generation apparatus 100 is configured by installing a computer program for realizing each functional block illustrated in a general-purpose personal computer and preparing each database. For example, the database may be provided online from a server, or may be provided by a recording medium such as a CD-ROM. In addition, in the present embodiment, each functional block described below is configured by software by the CPU of the personal computer executing a computer program as described above, but may also be configured by hardware. It is.

  Each database in the figure used in the present embodiment will be described. The altitude data 140 is a database that stores altitude values in association with each mesh defined by dividing the ground surface into fixed sections. As an example, data published by the Geographical Survey Institute can be used. In this case, the altitude does not necessarily indicate an average value or a maximum value in the mesh, but merely means that any one point in the mesh has this altitude value.

  The city map data 142 is two-dimensional map data in which roads, buildings, other features, and terrain are represented by polygons. Attributes such as features are also stored in association with the polygons. For example, for roads, road types such as national roads and prefectural roads, the number of lanes, traffic regulations, and the like are stored as attributes.

  The photograph data 144 is image data obtained by photographing the terrain from a high place. In this embodiment, so-called aerial photographs are used. The photograph data 144 is data representing a planar shape in a coordinate system unified with the city map data 142 by a transformation process such as affine transformation.

  The road network data 146 is data that can be used for route search, and represents the state of road connection as a set of links and nodes. In the road network data 146, when roads intersect on the same plane, nodes are always set at the intersections. In the case of a public intersection in a different plane such as a three-dimensional intersection, no node is set.

  The feature three-dimensional data 148 stores a three-dimensional model representing the shape of each feature represented in the three-dimensional map data. For example, a standard three-dimensional model is stored for standard features such as street trees and signals. For a building having a characteristic shape, a three-dimensional model generated individually is stored. Since various known techniques can be applied to these three-dimensional model generation methods, detailed description thereof is omitted here.

  The texture data 150 stores image data pasted on the surface of each polygon in the 3D map data. The texture data includes, for example, an image representing the appearance of a building, an image representing the road surface and the sea surface, and an image representing a signboard.

  In the generation apparatus 100, a command input unit 102 inputs various commands through user operations on a keyboard, a mouse, and the like. When the user designates a region to be processed and inputs a 3D map data generation command, the 3D map data generation unit 110 uses each of the function blocks shown below while using the various databases described above. Are linked to generate 3D map data. The output unit 104 outputs the generated 3D map data to a server or the like via a network, or writes it on a recording medium such as a CD-R.

  In the present embodiment, in the process of creating 3D map data, six types of processes are executed: ground polygon generation processing, digging processing, three-dimensional intersection processing, elevated form identification processing, tunnel processing, and finishing processing. The ground surface polygon generation process is a process for generating three-dimensional polygon data of the ground surface having undulations in consideration of the altitude. This process is executed by the ground polygon generation unit 120 using the altitude data 140.

  The digging process is a process for generating three-dimensional polygon data of topography and features having a surface cut off substantially vertically, such as a pier at a port or a platform of a station. This processing is executed by the digging processing unit 122 using the altitude data 140 and the city map data 142.

  The three-dimensional intersection process is a process for specifying an intersection type such as an elevated, embankment, and cut for a place where a road or a railway is three-dimensionally intersected, and generating a three-dimensional model reflecting these formats. This process is executed by the three-dimensional intersection processing unit 124 using the city map data 142 and the photograph data 144.

  Elevated form identification processing is a method of generating a 3D model by selecting either a parallel form in which the vertical lines are parallel or a vertical form in which the vertical lines are superimposed vertically on the elevated road. It is a process to do. This process is executed by the elevated form processing unit 126 using the city map data 142 and the photograph data 144.

  The tunnel processing is processing for specifying a location where a tunnel is provided and generating a three-dimensional model. This processing is executed by the tunnel processing unit 128 using the city map data 142 and the road network data 146.

  The finishing process is a process for completing the three-dimensional map data by using the result of each of the processes described above and arranging a building or the like and pasting a texture. The finishing process is executed by the following functional blocks. The feature placement unit 130 executes a process of placing each three-dimensional model stored in the feature three-dimensional data. The roadside tree arrangement unit 132 arranges a 3D model of the roadside tree in the section designated by the user. The texture pasting unit 134 pastes the texture stored in the texture data 150 on the surface of the three-dimensional model. In creating the three-dimensional map data, some of the functional blocks exemplified here may be omitted or further functional blocks other than those exemplified may be provided according to the required accuracy.

B. Ground polygon generation processing:
FIG. 2 is a flowchart of the surface polygon generation process. This process is executed by the CPU of the generation apparatus 100 and corresponds to the function of the ground polygon generation unit 120. When this process is started, the CPU first inputs elevation data (step S1). An example of elevation data is shown schematically in the figure. As described above, the elevation data is data that gives the height P1 for each mesh M1. On the basis of the height P1, the mesh surface defined in the xz plane is translated in the height direction, whereby a stepped ground polygon M2 shown in the figure can be generated.

The CPU then defines a triangular polygon that passes through each center of gravity, assuming that the height P1 represents the height at the center of gravity of each mesh (step S2). By doing so, a ground polygon GS having continuous undulations is formed as shown.

C. Mining treatment:
FIG. 3 is an explanatory diagram showing an outline of the digging process. In the upper part of the figure, a two-dimensional map including a port pier is illustrated. The white part is the land, and the hatched part is the sea surface. Dashed lines are meshes that provide elevation data. Here, a case will be described in which a region corresponding to the sea surface is dug down to express a terrain having a vertical cut-off side like a pier.

  In this two-dimensional map, the elevation distribution along the line AA is shown in the middle of the figure. Since the land and the sea surface are mixed in the mesh a1, the altitude does not necessarily represent the value of the land portion. Even if the altitude is a value corresponding to the sea level, it is not always 0 m above sea level. Here, it is assumed that a relatively low altitude value equivalent to the sea level is set as indicated by a broken line. Since most of the meshes a2, a3, and a5 are the sea surface, similarly, it is assumed that relatively low elevation values are set. The mesh a4 is assumed to have a relatively high elevation value corresponding to the jetty surface, as indicated by the solid line in the figure.

  In this example, the sea surface polygon is the target polygon for the mining process. In the present embodiment, prior to the excavation process, the altitude value of the mesh including the land among the meshes adjacent to the mesh corresponding to the sea surface is given. If the adjacent mesh is also sea level, this process is omitted. In this way, among meshes corresponding to the target polygon, a mesh whose adjacent mesh includes land is called a boundary mesh.

  As an example of the processing described above, for example, as shown in the upper diagram, the altitude value of one of the meshes b1 and c1 is given to the mesh a1. Similarly, the altitude value of mesh b2 is given to mesh a2, mesh c3 to mesh a3, and mesh a4 to mesh a5. Adjacent meshes can be selected by an arbitrary method. For example, the altitude value of the mesh a4 may be given to the mesh a3. As a result, as shown in the middle diagram, the altitude values of the meshes a1 to a5 are corrected as indicated by the arrows, and become constant altitude values as indicated by hatching.

  Next, from this state, the part corresponding to the polygon on the sea surface is dug down. That is, the area corresponding to this polygon is set to an altitude value of zero. As a result, as shown in the lower part of the figure, on the AA line, the altitude value is 0 except for the portion corresponding to the pier, and a vertical cut surface appears in the portion corresponding to the pier. .

  The broken line in the lower figure represents the ground surface when the above-described digging process is not performed. It can be seen that the portion corresponding to the pier may not be properly represented if the digging process is not performed. Further, when the above-described elevation value correction is not performed, even if only the portion corresponding to the sea level is forcibly set to the elevation value 0, as shown by the broken line B, it is unrealistic on the upper surface of the pier. An edge will appear. The digging process of the present embodiment can avoid such an adverse effect and appropriately model a topography having a vertical cut-off.

  FIG. 4 is a flowchart of the digging process. This process is executed by the CPU of the generation apparatus 100 and corresponds to the function of the digging processing unit 122. When this process is started, the CPU reads elevation data and city map data (step S10). Then, based on the city map data, a target polygon to be mined is selected (step S12). As the target polygon, a method of selecting a sea surface, a river, a laid track in a station premises, etc. based on attributes can be adopted.

  Next, the CPU selects a boundary mesh from among the meshes included in the target polygon, and corrects the elevation (step S14). The correction method is as described above with reference to FIG. The altitude may be corrected for all meshes included in the target polygon, not limited to the boundary mesh.

  When the correction of the altitude is completed, the CPU digs up a portion corresponding to the target polygon (step S16). In the present embodiment, either one of the two excavation methods, absolute excavation and relative excavation, is selected and executed. Absolute digging is a process of digging down so that the surface of the polygon has a specified elevation value AH. The example described in FIG. 3, that is, the process of digging the sea surface portion to an altitude value of 0 corresponds to this. The relative excavation is a process of excavating the surface of the polygon by a value DH specified from the elevation value before the excavation. This process is suitable when the target polygon is a laying track in a river or a station. By specifying the digging method in advance in association with the attribute of the target polygon, it is possible to realize the use of absolute digging and relative digging relatively easily. The CPU repeatedly executes the above processing until all the target polygons are completed (step S18).

  In a present Example, after performing the above-mentioned digging process, the connection process and bridge installation process which are mentioned later are further performed. The linking process is a process for continuously connecting the polygons when elevation mismatch occurs between polygons whose elevation should be continuously changed as a result of the mining process. For example, the present invention can be applied when an elevation mismatch occurs between a river polygon and a sea surface polygon. The bridge installation process is a process for determining the height of the bridge over the river. As described above, since the elevation data is given based on a mesh set regardless of the bridge position, the height of the bridge is not necessarily given. Bridges are usually hung between embankments at both ends of the river. In the bridge installation process, considering these circumstances, the height of the levee is obtained based on the height distribution at both ends of the river, and the height of the bridge is set. The connection process and the bridge installation process may be omitted according to an instruction from the user.

  FIG. 5 is a flowchart of the connection process. The CPU of the generation apparatus 100 is a process executed after the digging process (FIG. 4) and corresponds to the function of the digging process unit 122. When this process is started, the CPU reads the ground surface data generated by the digging process (step S20). Next, an adjacent polygon adjacent to the target polygon subjected to the drilling process is selected (step S22), and the necessity of the connection process is determined (step S24). In this embodiment, when the target polygon and the adjacent polygon have the same attribute, it is determined that the connection process is not necessary, and when both are different, it is determined that the connection process is necessary.

  When determining that the connection process is necessary, the CPU generates a connection polygon for continuously connecting the altitudes of the target polygon and the adjacent polygon (step S26). In the figure, a method for generating a connected polygon is illustrated. It is assumed that the polygon P2 located on the right side of the figure is a target polygon, for example, the sea surface. It is assumed that the polygon “P1 + P31” located on the left side of the figure is an adjacent polygon, for example, a river. As shown in the drawing, there is a discontinuous elevation difference H1 between the two at the beginning.

  In the present embodiment, the adjacent polygon P32 is divided into polygons P1 and P31 at a vertex Ep1 that is relatively close to the target polygon P2, and the elevation of some vertices of the polygon P31 is corrected, thereby connecting the connected polygon P32 that connects them. Set. In this embodiment, the polygons P1 and P32 are defined as separate polygons, but they may be combined and defined as one polygon.

  With respect to the polygon P32, the shape of the side E1 connecting the vertex Ep1 connected to the polygon P1 and the vertex Ep2 connected to the polygon P2 can be variously set. As shown by the thin solid line on the right side of the figure, the altitude may be set so as to change linearly from the vertex Ep1 to Ep2, or as shown by the thick line, it changes with a curve, particularly a Bezier curve or a spline curve. You may set as follows. From the viewpoint of smoothly connecting the polygons P1 and P2, it is preferable to set the polygons in a curved shape. The CPU executes the above processing for all polygons (step S28).

  FIG. 6 is a flowchart of the bridge installation process. The CPU of the generation apparatus 100 is a process executed after the digging process (FIG. 4) and corresponds to the function of the digging process unit 122. Either the bridge installation process or the connection process (FIG. 5) may be executed first.

  When this process is started, the CPU reads the ground surface data generated by the digging process (step S30). Next, referring to the attribute of the city map data, a bridge polygon is selected (step S32), and the bridge height is set (step S34). An example of setting the bridge height is shown in the figure. The city map data is shown above. In this example, the bridge Brg is straddling the river R3, and roads Ro1 and Ro2 are connected to both ends thereof. On both sides of the river R3, there are riverbanks R2, R4 and dikes R1, R5. Roads Ro1 and Ro2 are uphill from both ends toward embankments R1 and R5, respectively. The bridge is hung at the height of embankments R1 and R5.

  In the lower row, the height distribution along the straight line Li is shown. The broken lines in the figure represent mesh divisions. River R3 and Kawahara R4 have relatively low elevation values, and banks R1 and R5 have relatively high values. In the present embodiment, the height of the bridge Brg is made to coincide with the highest value among the height distributions obtained along the road including the bridge Brg. By carrying out like this, as shown to the continuous line B2 in a figure, the bridge over the embankment is realizable. If the height of the bridge Brg is set in the same way as the ground surface polygon based on the height distribution, it becomes an unreal state as shown by the broken line B1, but according to the method of this embodiment described above, It is possible to avoid such a situation and realize an appropriate bridge model. The CPU executes the above processing for all the bridges (step S36).

D. Three-dimensional intersection processing:
FIG. 7 is an explanatory diagram showing an outline of the solid intersection process. There are three types of road and railroad crossings: elevated, embankment, and cut. The elevated means that the upper road OP is supported by the pier BP. The lower road UP often exists on the ground plane. The embankment is a form in which the upper road OP is formed on the upper surface where the soil is raised. The lower road UP is located on the ground plane and is often provided so as to pass through a tunnel provided in the embankment. The cut is a form in which the lower road UP passes through a groove-shaped portion formed by cutting the ground plane. The upper road is located on the ground plane, and the cut portion is passed by the bridge Brg1.

  In the present embodiment, the above three types are discriminated based on the state of the shadow near the three-dimensional intersection and the position of the nearby building. As shown, in the case of an elevated road, the shadow S1 of the upper road OP can be outside the road on at least one side of the road. The same applies to the embankment. However, unlike the elevated shadow S1, the embankment shadow S2 often has a wavy boundary.

  Elevations and embankments can be identified by nearby buildings. For example, in the case of an elevated structure, the space below the upper road OP is often vacant. Therefore, when viewed in plan, the distance d1 between the building and the upper road OP is often relatively small. . In the case of embankment, the distance d2 between the building and the upper road OP is often relatively large because the upper road OP is buried under the soil. Therefore, for example, it is possible to distinguish between the elevated and the embankment based on the magnitude relationship between the distance d between the building and the upper road OP and the threshold value Th arbitrarily set in a range larger than d1 and smaller than d2. it can. That is, if d <Th, it can be determined that it is elevated;

  On the other hand, in the case of cut, the shadow S3 of the lower road UP does not protrude from the road. Further, the shadow S4 of the upper road OP does not protrude from the lower road UP.

  FIG. 8 is an explanatory diagram showing an example of cutting determination. A cut photo is shown as an example. In this photograph, roads R11 and R12 pass through the upper side of the road R13 passing through the lower side. The shadows S13, S11, S12 of the roads R13, R11, R12 do not protrude from the road R13. Therefore, it can be determined from these shadow states that the road R13 is cut.

  FIG. 9 is a flowchart of the solid intersection process. This process is executed by the CPU of the generation apparatus 100 and corresponds to the function of the solid intersection processing unit 124. When this process is started, the CPU reads city map data and photograph data (step S40). Then, based on the city map data, the location of the three-dimensional intersection is extracted (step S42).

  In the figure, an example of extraction of the location of the three-dimensional intersection is shown. The example in the figure shows a part of the city map data, and shows a state in which roads R41 to R44 intersect at C1 to C4, respectively. At the intersection C2, since the boundary line of the roads R42 and R44 is not drawn, it is determined that the intersection is a normal intersection. The intersections C1, C3, and C4 are determined to be solid intersections in which the road on which the boundary is drawn passes above and the road on which the boundary is divided passes below. For example, the intersection C1 is a three-dimensional intersection passing through the road R43 on the upper side and the road R42 on the lower side. A three-dimensional intersection can be extracted based on the presence / absence of a road boundary line and the division status.

  Next, the CPU specifies the format of these three-dimensional intersections based on the conditions described above with reference to FIG. 7 (step S44). Various known image processing techniques can be applied to determine whether the shadow protrudes from the road or to determine the distance between the building and the road. The specification of the format may be performed completely automatically, or semi-automatically, that is, the user specifies the location of the shadow or the position of the building with the pointer, and determines based on this specification. Also good.

  In this embodiment, after determining the form of the three-dimensional intersection, an inclination adjustment process for smoothly changing the altitude along the road corresponding to the elevated, embankment, and cut is executed. In the tilt adjustment process, check points are used. The check point is a point where the altitude of the road can be almost determined, and examples thereof include a point where a three-dimensional intersection occurs and a break point where the road type changes from cut to elevated or embankment.

  FIG. 10 is an explanatory diagram showing an example of checkpoints. Photo data near the road RO is shown in a simplified manner. This road RO passes over the roads UP1 and UP2 on the ground plane with an elevated route, and then passes under the roads OP1 and OP2 on the ground plane with cuts. These road passing points, P51, P52, P54, and P55, are checkpoints.

  The breakpoint is determined by the following procedure. From the example of the figure, the shadow S51 of the road RO can identify the section D1 formed outside the road RO as the elevated section. Further, the section D2 in which the shadow S52 is within the road RO can be specified as the cut section. The break point is a point corresponding to the boundary between the sections D1 and D2. Therefore, in this embodiment, the midpoint P51 of the gap between the sections D1 and D2 is used as a breakpoint. For example, in the example of the figure, when the elevated section D1 and the cut section D2 cannot be specified from the shadows S51 and S52, the midpoint of the checkpoints P52 and P54 may be set as the breakpoint. In the present embodiment, as an inclination adjustment process, the altitude of the road RO is given at these check points, and the altitudes of the respective points are set by connecting these check points with spline curves.

  FIG. 11 is a flowchart of the inclination adjustment process. The CPU of the generation device 100 is a process executed after the solid intersection process (FIG. 9) and corresponds to the function of the solid intersection processing unit 124. When this process is started, the CPU reads the city map data and the result of specifying the three-dimensional intersection format (step S50), and selects a road to be processed (step S51). Then, along this road, a check point is specified and its altitude is set (steps S52 and S53). In this embodiment, the elevation value of the three-dimensional intersection at the elevated or embankment on the road on the ground plane is a value obtained by adding a specified value H, for example, 11 m to the height of the ground plane. The elevation value of the three-dimensional intersection passing through the cut under the road on the ground plane was a value obtained by subtracting the specified value H from the height of the ground plane. In addition, the elevation value of the breakpoint was the ground plane height.

  The CPU obtains the altitude of each other point by connecting the altitude values of the check points set in this way with a spline curve (step S54). An example of the tilt adjustment process is shown in the figure. This figure shows the elevation distribution along the road. Before the processing, it is assumed that the points Pa1, Pb1, and Pc1 that do not correspond to the check points are set to the altitude value H corresponding to the overhead. As described above, the altitude value is set to H at a point where it is clear that the road to be processed is an elevated point. At a point that is clearly cut, the altitude value is set to “−H”. At the breakpoint, the elevation value is set to zero. The CPU obtains a height distribution indicated by a solid line by obtaining a spline curve that passes through each of these points. The elevation values of the points Pa2, Pb2, and Pc2 on the spline curve are given to the points Pa1, Pb1, and Pc1 on the road. By doing so, the height of the road can be changed smoothly, and a realistic three-dimensional model of the road can be created. The CPU executes the above processing until all the roads are completed (step S55).

E. Elevated form specific processing:
FIG. 12 is an explanatory diagram showing an outline of the elevated form specifying process. There are two types of elevated types: parallel type and vertical type. As shown in the lower part of the figure, the parallel type refers to a road in which upper and lower lines are arranged in parallel on the left and right. The vertical type refers to a road in which vertical lines are arranged vertically as shown in the upper part of the figure. The arrangement of the uplink and the downlink may be opposite to the illustrated example.

  Judgment of elevated type is performed according to the following procedure. A map display example for the vertical elevated road is shown on the left side of the middle of the figure, and a photo example is shown on the right side. As shown in the figure, map data such as city map data and road network data are often deviated from each other so that vertical lines can be visually distinguished even on a vertical road. In contrast, in the photo data, only one of the lanes is shown. In the present embodiment, in this way, the part of the elevated road described in the map data that is divided into the upper and lower lines is determined to be a vertical road, and the other part is determined as the road where both lanes cannot be confirmed in the photograph data. Judged as parallel type. For example, when the difference between the road width specified based on the map data and the road width specified from the photo data is equal to or greater than a predetermined value, the map data and the photo data do not match, that is, a method of determining a vertical road. Can be taken. As described above, by using the map data and the photograph data together, it is possible to generate a three-dimensional model in an elevated form that matches the reality.

  FIG. 13 is a flowchart of the elevated form specifying process. This process is executed by the CPU of the generation apparatus 100 and corresponds to the function of the elevated form processing unit 126. When this process is started, the CPU reads city map data and photograph data (step S60). Instead of the city map data, road network data may be used. Based on the city map data, an arbitrary target elevated section to be processed for specifying the elevated form is selected (step S61), and the number of lanes in the target elevated section is identified from the city map data and the photograph data ( Step S62). If the two match, the CPU generates a three-dimensional model in a parallel elevated form, and if not, generates a three-dimensional model in a vertical elevated form (steps S63 to S65). For example, in the processes of steps S64 and S65, a basic three-dimensional model can be prepared in advance for each of the parallel type and the vertical type, and this three-dimensional model can be arranged in the target elevated section. The CPU executes the above processing until all the roads are completed (step S66).

F. Tunnel processing:
FIG. 14 is a flowchart of tunnel processing. The tunnel processing is processing for specifying a location where a tunnel is provided and generating a three-dimensional model. This process is executed by the CPU of the generation apparatus 100 and corresponds to the function of the tunnel processing unit 128. When this process is started, the CPU reads road network data and city map data (step S70). Then, any arbitrary link is selected as a processing target from the road network data (step S71). Next, in the city map data, it is determined whether or not the portion corresponding to this link has been subjected to hidden line processing (step S72). An example of hidden line processing is shown on the right side of the figure. In the city map data, the road Rb1 is usually represented by a polygon composed of a solid boundary line as shown in the figure, and is represented by a broken line in the tunnel section. There is a case where a description TN indicating a wellhead is given at the entrance of the tunnel. As described above, when the portion corresponding to the road link is hidden line-processed by the city map data, the CPU determines that the section is a tunnel (step S72), and generates a three-dimensional model of the tunnel (step S72). S73). In addition to the presence / absence of hidden line processing, the determination condition may include the presence / absence of a description TN representing a wellhead. For this process, for example, a basic three-dimensional model representing the shape of the tunnel is prepared in advance, and a method of arranging this for the tunnel section can be adopted. The CPU executes the above processing until all the roads are completed (step S74).

G. Finishing process:
FIG. 15 is a flowchart of the finishing process. This process is executed by the CPU of the generation apparatus 100 and corresponds to the functions of the feature arrangement unit 130, the roadside tree arrangement unit 132, and the texture pasting unit 134. When this process is started, the CPU reads three-dimensional ground surface data and city map data (step S80). The three-dimensional ground surface data means a three-dimensional model generated by a ground polygon generation process, a mining process, a three-dimensional intersection process, an elevated form specifying process, and a tunnel process. As a result of these processes, the three-dimensional ground surface data represents general roads including undulating ground surface, sea surface, river topography, and elevated roads and tunnels.

  The CPU arranges a three-dimensional model of the feature with reference to the city map data with respect to the three-dimensional ground surface data (step S81). The three-dimensional model of the feature is stored in the feature three-dimensional data 148 as described above. A three-dimensional model of the feature may be newly generated during the process of step S81.

  When the user designates a street tree arrangement section, the CPU inputs this designation (step S82), and generates a street tree model (step S83). In this embodiment, a basic three-dimensional model of a roadside tree is prepared in advance, and the height and arrangement are changed based on 1 / f fluctuation. An example of arrangement of street trees is shown in the figure. It is assumed that the height of the basic model is H1, and a straight line in the figure is designated as the arrangement section. The CPU expands and contracts the basic model to change the height and change the deviation in the direction intersecting the straight line. As a result, the second street tree has a height H2 different from that of the basic model, and is arranged at a position including the deviation Ep from the straight line. About arrangement | positioning, you may shift about not only the direction which cross | intersects a straight line but the direction along a straight line. In this way, a natural street tree model can be realized by shifting the height and arrangement. The height and arrangement may be determined using a random number regardless of 1 / f fluctuation.

  Next, the CPU pastes a texture using the texture data 150 (step S84). In the drawing, an example in which the texture TEX is pasted on the building polygon POL is shown. As described above, by attaching an appropriate texture to the surface of each polygon, a realistic three-dimensional model can be generated. Here, the building polygon POL is illustrated, but various textures such as a road surface, a sidewalk, and a signboard can be pasted.

  According to the generating apparatus 100 of the present embodiment, more realistic 3D map data can be generated with a relatively light load by the processes described above. It is not always necessary to perform all of the processes described in the embodiment, and some processes may be omitted as appropriate based on a request for accuracy of the three-dimensional map data. The processes described in the embodiments are semi-automatic, that is, a part of these processes may be executed based on designation by a user, command input, and the like. As mentioned above, although the various Example of this invention was described, it cannot be overemphasized that this invention is not limited to these Examples, and can take a various structure in the range which does not deviate from the meaning.

It is explanatory drawing which shows schematic structure of the production | generation apparatus of the three-dimensional map data in an Example. It is a flowchart of a surface polygon generation process. It is explanatory drawing which shows the outline | summary of a digging process. It is a flowchart of a digging process. It is a flowchart of a connection process. It is a flowchart of a bridge installation process. It is explanatory drawing which shows the outline | summary of a three-dimensional intersection process. It is explanatory drawing which shows the example of judgment of cut. It is a flowchart of a three-dimensional intersection process. It is explanatory drawing which shows the example of a checkpoint. It is a flowchart of an inclination adjustment process. It is explanatory drawing which shows the outline | summary of an elevated form specific process. It is a flowchart of an elevated form specific process. It is a flowchart of a tunnel process. It is a flowchart of a finishing process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... Generation apparatus 102 ... Command input part 104 ... Output part 110 ... Three-dimensional map data generation part 120 ... Ground polygon generation part 122 ... Underground processing part 124 ... Three-dimensional Intersection processing unit 126 ... Elevated form processing unit 128 ... Tunnel processing unit 130 ... Feature placement unit 132 ... Street tree placement unit 134 ... Texture pasting unit 140 ... Elevation data 142 .. .Map data 144 ... Photo data 146 ... Road network data 150 ... Texture data

Claims (19)

A generating device for generating 3D map data,
An altitude data reference section for referring to altitude data given to a mesh defined by dividing the ground surface into predetermined areas;
A map data reference section for referring to polygon data representing terrain and two-dimensional map data including its attribute data;
Among the polygons in the map data, a target polygon selection unit that selects a target polygon to be a target of the mining process,
Altitude data given to a mesh outside the region of the target polygon located in the vicinity thereof is assigned to each mesh including the boundary line of the target polygon, and each mesh including the boundary line of the target polygon ; and elevation data assigned to each mesh, based on the ground surface data generating unit that the generating a three-dimensional data representing the elevation of the ground surface contained in each mesh,
A digging processing unit that executes digging processing for correcting the elevation value of the region corresponding to the target polygon to a predetermined value set regardless of the elevation data for the three-dimensional data on the ground surface apparatus. The generating device according to claim 1,
The target polygon is a polygon included in an area representing the sea surface in the map data ,
The generating device is a generating device in which the predetermined value is a preset fixed value. The generating device according to claim 1,
The target polygon is a polygon included in an area representing a river in the map data ,
The generation device is provided with the predetermined value given as a predetermined relative value with respect to an altitude near the target polygon. The generating device according to claim 1,
The mining processing unit
A selection unit that selects, as a connected polygon, a polygon whose elevation should be continuously changed from the target polygon among polygons adjacent to the target polygon;
After the digging process, at the connection portion between the target polygon and the connected polygon , (i) the target polygon or a part of the connected polygon is deformed, and the altitude of the target polygon and the connected polygon Or (ii) dividing the target polygon or a part of the connected polygon to obtain the altitude of the target polygon and the height of the connected polygon. A generation apparatus including a connection processing unit that secures continuity of elevation between the target polygon and the connected polygon by generating a connected polygon including a vertex having an altitude to be complemented . The generating device according to claim 1 , further comprising:
A photo data reference section for referring to photo data obtained by photographing an area corresponding to the map data from a high place;
Based on the map data, an extraction unit that extracts a plurality of three-dimensional intersections as processing targets;
Based on the photograph data, a positional relationship between the traffic road and at least one of a shadow of the traffic road to be processed and a shadow formed on the traffic road is specified, and the traffic to be processed is determined based on the positional relation. A generation apparatus comprising: an intersection type specifying unit that specifies an intersection type of a road. The generating device according to claim 5,
The said intersection type specific | specification part is a production | generation apparatus which judges that it is an overpass or embankment about the traffic road in which the shadow of self protrudes among the said process objects. The generating device according to claim 6, wherein
The intersection type specifying unit further refers to the map data, and among the traffic roads determined to be the elevated or embankment, the intersection type specifying unit is based on a distance between a building adjacent to the traffic road and the traffic road. And a generator for distinguishing fills. The generating device according to claim 5,
The said intersection type specific | specification part is a production | generation apparatus which judges that it is cut with respect to the traffic route in which the shadow of an inside is settled among the said process objects. The generating device according to claim 1 , further comprising:
A traffic network data reference unit for referring to traffic network data that represents the connection of traffic routes in terms of links and nodes;
A generation apparatus comprising: a tunnel specifying unit that specifies that a tunnel is provided in a portion represented by a broken line in the map data for a traffic route corresponding to a link of the traffic route network data. The generating device according to claim 1 , further comprising:
A photo data reference section for referring to photo data obtained by photographing an area corresponding to the map data from a high place;
Wherein by comparing the map data and the photographic data, generating apparatus comprising a vertical line number is different traffic channel, the vertical lines and elevated type identifying unit that identifies that the vertical elevated path provided in different planes. The generation apparatus according to claim 10, comprising:
The elevated format specifying unit on the basis of the difference of the traffic channel width in the map data and the photographic data, generator for determining the difference of the vertical line number. A generation method for generating 3D map data by a computer,
As a step executed by the computer,
Referring to elevation data given to a mesh defined by dividing the ground surface into predetermined regions;
A step of referring to polygon data representing terrain and two-dimensional map data including attribute data thereof;
A step of selecting a target polygon to be a target of the mining process among the polygons in the map data;
Altitude data given to a mesh outside the region of the target polygon located in the vicinity thereof is assigned to each mesh including the boundary line of the target polygon, and each mesh including the boundary line of the target polygon ; and elevation data assigned to each mesh, based on the steps of generating three-dimensional data representing the elevation of the ground surface to be included in each mesh,
And a step of executing a digging process for correcting the elevation value of the region corresponding to the target polygon to a predetermined value set regardless of the elevation data for the three-dimensional data on the ground surface. The generation method according to claim 12 , comprising:
As a step executed by the computer ,
A step of referring to photograph data obtained by photographing a region corresponding to the map data from a high place;
Extracting a plurality of three-dimensionally intersecting traffic paths as processing targets based on the map data;
Based on the photograph data, a positional relationship between the traffic road and at least one of a shadow of the traffic road to be processed and a shadow formed on the traffic road is specified, and the traffic to be processed is determined based on the positional relation. And a step of specifying a road intersection type. The generation method according to claim 12 , comprising:
As a step executed by the computer ,
Referencing the traffic network data representing the connection of the traffic route with links and nodes;
A generation method comprising: a step of specifying that a tunnel is provided in a portion represented by a broken line in the map data for a traffic route corresponding to the link of the traffic route network data. The generation method according to claim 12 , comprising:
As a step executed by the computer ,
A step of referring to photograph data obtained by photographing a region corresponding to the map data from a high place;
Wherein by comparing the map data and the photographic data, generating method comprising the step of vertically ruling different traffic channel, the vertical line is identified as a vertical elevated path provided in different planes. A computer program for generating 3D map data,
A function for referring to elevation data given to a mesh defined by dividing the ground surface into predetermined areas;
A function for referring to polygon data representing terrain and two-dimensional map data including its attribute data,
A function of selecting a target polygon to be a target of the mining process among the polygons in the map data;
Altitude data given to a mesh outside the region of the target polygon located in the vicinity thereof is assigned to each mesh including the boundary line of the target polygon, and each mesh including the boundary line of the target polygon ; and elevation data assigned to each mesh, based on a function of generating three-dimensional data representing the elevation of the ground surface to be included in each mesh,
To cause a computer to execute a digging process for correcting the elevation value of an area corresponding to the target polygon to a predetermined value set regardless of the elevation data for the three-dimensional data on the ground surface Computer program. The computer program according to claim 16 , further comprising:
A function for referring to photograph data obtained by photographing an area corresponding to the map data from a high place;
Based on the map data, a function of extracting a plurality of traffic paths intersecting three-dimensionally as processing targets;
Based on the photograph data, a positional relationship between the traffic road and at least one of a shadow of the traffic road to be processed and a shadow formed on the traffic road is specified, and the traffic to be processed is determined based on the positional relation. A computer program for causing a computer to realize a function for specifying a road intersection type. The computer program according to claim 16 , further comprising:
A function to refer to the traffic network data that represents the connection of the traffic route with links and nodes,
A computer program for causing a computer to realize a function of specifying that a tunnel is provided in a portion represented by a broken line in the map data for a traffic route corresponding to the link of the traffic route network data. The computer program according to claim 16 , further comprising:
A function for referring to photograph data obtained by photographing an area corresponding to the map data from a high place;
Wherein by comparing the map data and the photographic data, the vertical line number is different transportation path, a computer program for the upper and lower lines to realize the function of identifying to the computer as a vertical elevated path provided in different planes.

Images