JP5946369B2 - 3D map image data generation system - Google Patents

Description

  The present invention relates to a three-dimensional map image data generation system that generates data for drawing a three-dimensional map that represents a feature three-dimensionally.

A three-dimensional map in which features such as buildings and roads are three-dimensionally displayed is variously used in the form of being displayed on a display in the form of a car navigation system, an electronic map, or the like, or in the form of printed matter. A three-dimensional map is usually drawn sequentially by a method such as perspective projection based on a three-dimensional model representing the shape of a building or the like. Patent Document 1 also provides a technique for preparing two-dimensional image data drawn in parallel by projecting a three-dimensional model in advance and displaying a three-dimensional map with a light load without using the three-dimensional model at the time of display. It is disclosed.
These 3D maps have the advantage that it is easier for the user to intuitively grasp the current position and geographical information because the situation of buildings and the like can be drawn in a state close to reality compared to a 2D map. is there. Accordingly, it is desired to provide a more realistic map for a three-dimensional map in order to improve the intuitive recognition of such a user.

  As a method for improving the reality in a three-dimensional map, use of a texture created based on an image or the like of a real building is known. Patent Document 2 discloses a specific method for improving the reality by applying a texture to the surface of a three-dimensional model.

JP 2011-186960 A Japanese Patent No. 4812011

As disclosed in Patent Document 1, when a three-dimensional map viewed from a position higher than a building is drawn, the roof surface of the building is relatively conspicuous in both parallel projection and perspective projection. In the past, attention has not been paid to improving the reality of the rooftop. If the roof surface is simply drawn in a planar shape, the reality of the entire three-dimensional map may be impaired, and the value of the three-dimensional map may be reduced.
As a method for solving these problems, it is possible to improve the reality of the roof surface using texture as in the prior art. However, in order to create the texture of the roof surface, it is necessary to photograph the building from a position where the roof surface of the building is visible, which is not easy.
Further, if the operator prepares the image of the roof individually by retouching the image or the like, the work load becomes great.
In view of such an excessiveness, an object of the present invention is to improve the reality of a rooftop surface with a relatively light load in a three-dimensional map.

The present invention
A 3D map image data generation system for generating 3D map image data for rendering a 3D map representing a building in 3D,
A 3D map database storage unit storing a 3D model representing the 3D shape of the building;
A projection processing unit that projects the three-dimensional model by a predetermined projection method to generate a normal projection image of a three-dimensional map;
Using the image of the area corresponding to the rooftop of the building among the normal projection images generated by the projection processing unit, the rooftop texture of each building is subjected to image processing for giving a sense of unevenness based on a predetermined rule. A roof texture generator for generating
The roof texture can be combined with the normal projection image to form a 3D map image data generation system including an image composition unit that generates the 3D map image data.
The generated three-dimensional map image data can be used for drawing a three-dimensional map in a manner such as display on a display or print output by a printing apparatus. Since the three-dimensional map image data is data as a two-dimensional image generated by the projection process, it is not necessary to perform the projection process again for display, print output, and the like.

According to the present invention, the normal projection image generated based on the three-dimensional model is subjected to image processing based on a predetermined rule, thereby giving a sense of unevenness using an image of an area corresponding to the rooftop. be able to. This image processing is based on a predetermined rule set in advance and can be performed by automatic processing, so that it does not impose a great processing load on an operator or the like. However, in addition to such automatic image processing, pre-processing or post-processing by an operator may be performed.
Further, the buildings to be processed may be narrowed down according to the attribute. For example, it is conceivable to narrow down by a method such as a building having a predetermined height or higher, a predetermined type of building such as a building, or a building having a flat roof surface.

The unevenness imparted by the above method does not necessarily reproduce the appearance of the actual rooftop of a building. However, it is normal for the rooftop of a building to have various similar equipment such as walls or fences that prevent falling, drainage channels for distributing rainwater on the rooftop, and water tanks. Yes, it is possible to express these facilities by giving a feeling of unevenness, so that the reality of the rooftop surface can be improved even if it is not an accurate reproduction. And the reality of the whole three-dimensional map can be improved by improving the reality of a roof surface.
Conventionally, improvement of the reality in a three-dimensional map has been directed to reproducing the actual situation as much as possible, but the inventors of the present application need to accurately reproduce the actual situation for the reality of the rooftop surface. No, the present invention has been achieved by focusing on the fact that it should be expressed to the extent that the reality of the entire three-dimensional map is not impaired.

  The unevenness can be given by various methods. For example, image processing may be performed in a state where a wall or a fence with a predetermined width is provided around the roof surface. Further, the image processing may represent a state in which facilities such as a tank and a stairway entrance are arranged in a predetermined shape on the roof surface in a circular shape or a rectangular shape. Multiple types of image processing may be used selectively or in combination depending on the shape of the building, the area of the roof surface, and the like.

As an example of image processing, the following modes can be taken.
In the 3D map image data generation system of the present invention,
The roof texture generation unit may generate a roof top cut image obtained by cutting an edge of a region corresponding to the roof top, and generate the roof texture by a process of applying a shadow to the roof top cut image. .
By doing so, it is possible to express a state in which a wall having a predetermined width is provided around the roof surface. The cutting of the edge of the region can be performed by a method of cutting a predetermined width, a method of reducing the roof surface at a predetermined ratio, or the like.
When applying shading, it is preferable to determine the direction in which light and shade appear, assuming lighting equivalent to the lighting direction when projecting a three-dimensional map. By doing so, it is possible to apply a shadow without a sense of incongruity.

In the 3D map image data generation system of the present invention,
Furthermore, it has a boundary line generation unit that generates a boundary line image for attaching to the roof top surface by forming a contour line of a region cut out of the edge corresponding to the roof top surface,
The image composition unit may further compose the boundary line image.
By carrying out like this, the wall provided in the circumference | surroundings of a roof surface can be expressed clearly with an outline, and reality can be improved more. The contour line can be determined by a method of cutting a predetermined width, a method of reducing the roof surface at a predetermined ratio, or the like.

In the 3D map image data generation system of the present invention,
The projection processing unit unifies the colors of the surfaces constituting each building in the three-dimensional model, generates the processing image by performing the projection in a lighting state from directly above, and in the processing image, A rooftop image is generated by extracting the area represented by the brightness higher than the surface as the rooftop,
The roof texture generation unit may specify a region to be subjected to the image processing in the normal projection image using the roof top image.
At the stage of the 3D model, it is easy to specify the rooftop of each building individually, but the image obtained by projecting the 3D model lacks the attribute of the rooftop, so specify the rooftop It will not be easy to do. According to the above aspect, since the rooftop surface has the highest brightness compared to the side surface or the like by lighting from directly above, the rooftop surface can be easily identified based on the difference in brightness even in the projected image. Can do.
In the said aspect, it is sufficient if the color of the surface which comprises each building is unified for every building. Different buildings may have the same color or different colors.

  In the present invention, the method for specifying the roof surface is not limited to the above-described embodiment. For example, in the three-dimensional model, when an attribute that can specify the polygon of the rooftop is attached, the surface other than the rooftop is made transparent or excluded from the drawing target, and only the rooftop is set as the projection target, You may take the method of obtaining the above-mentioned rooftop image. Further, even when such an attribute is not attached, the above processing may be performed by determining a surface where the normal line of each polygon is within a predetermined range from the vertical direction as the roof surface.

In the present invention, it is not always necessary to have all the above-described features, and a part of them can be omitted or combined as appropriate.
The present invention may also be configured as a 3D map image data generation method for generating data for drawing a 3D map by a computer, or as a computer program for causing a computer to execute such drawing. Also good. Moreover, you may comprise as a computer-readable recording medium which recorded such a computer program. Recording media include flexible disks, CD-ROMs, magneto-optical disks, IC cards, ROM cartridges, punch cards, printed matter on which codes such as bar codes are printed, computer internal storage devices (memory such as RAM and ROM), and A variety of computer-readable media such as an external storage device can be used.

It is explanatory drawing which shows the structure of a three-dimensional map image data generation system. It is a flowchart (1) of 3D map image data generation processing. It is a flowchart (2) of 3D map image data generation processing. It is a flowchart (3) of 3D map image data generation processing. It is a generation process example (1) of 3D map image data. It is a generation process example (2) of 3D map image data. It is a generation process example (3) of 3D map image data.

A. System configuration:
FIG. 1 is an explanatory diagram showing a configuration of a three-dimensional map image data generation system.
The three-dimensional map image data generation system 100 is configured by installing computer software that realizes various functions shown in a computer having a CPU, a RAM, and a ROM, and a three-dimensional map (sometimes referred to as a 3D map). Is a system for generating three-dimensional map image data for rendering. In this embodiment, an example in which the system is configured by a computer that operates in a stand-alone manner is shown, but a plurality of computers, servers, and the like may be connected via a network.
The three-dimensional map image data is image data as a two-dimensional image obtained by projecting a three-dimensional model representing a three-dimensional shape such as a building.
By using the 3D map image data generated by the 3D map image data generation system 100, a 3D map can be displayed on the smartphone 300 or other display. Further, the 3D map 201 can be printed by outputting the 3D map image data to the printer 200. Since the three-dimensional map image data is data as a projected two-dimensional image, it is not necessary to perform a projection process again in these map display and print output.

The contents of each functional block constituting the 3D map image data generation system 100 will be described. Each functional block shown in FIG. 1 is mainly configured by software as described above, but a part or all of the functional blocks may be configured by hardware.
The 3D map database 101 is a database used when creating 3D map image data, and stores 3D map data including polygon data representing a 3D shape of a feature including a building.
The projection processing unit 111 reads 3D map data in the 3D map database 101 and performs a projection process to generate a projection image, that is, a two-dimensional image that three-dimensionally represents a building or the like. In this embodiment, the projection process is performed by parallel projection, and the two-dimensional image is raster data. Projection processing by perspective projection may be performed.
The projection condition designating unit 110 designates conditions for performing projection processing. In this embodiment, “normal” projection conditions and “processing” projection conditions are prepared. “Normal” projection is a condition for projecting using 3D map data as it is. An image obtained by “normal” projection is stored as a normal projection image 102. The “processing” projection is a condition for performing a projection process for specifying the roof of the building in the normal projection image 102. The rooftop image identified using the “processing” projection results is stored as the rooftop image 103.
The rooftop texture generation unit 113 generates a texture image representing the unevenness of the rooftop surface by image processing using the normal projection image 102 and the rooftop surface image 103. The texture image generated in this way is stored as the roof texture 105.
The boundary line generation unit 112 generates a frame-like image corresponding to the boundary line of the rooftop texture 105 based on the rooftop surface image 103. This frame-shaped image is stored as the boundary line image 104.
The image synthesizing unit 114 synthesizes the roof texture 105 and the boundary line image 104 with the normal projection image 102 to generate an image with a sense of unevenness on the roof surface of each building in the normal projection image 102. The image data generated in this way becomes the 3D map image 106.
Hereinafter, the processing content for generating the 3D map image 106 will be specifically described.

B. 3D map image data generation processing:
2 to 4 are flowcharts of 3D map image data generation processing. This corresponds to a process executed in conjunction with each of the functional blocks shown in FIG. 1, and is a process executed by the CPU of the 3D map image data generation system 100 in terms of hardware.
When the process is started, the 3D map image data generation system 100 (hereinafter sometimes simply referred to as “system 100”) accepts designation of the drawing range, the line-of-sight direction, and the projection method from the operator (step S10). Other conditions according to the projection method may also be input. For example, in the case of parallel projection, it may be possible to input an azimuth and a look-down angle as the line-of-sight direction. In the case of perspective projection, it is conceivable to further input a viewpoint position.
Next, the system 100 reads 3D map data corresponding to the designated drawing range (step S12). And a normal projection image is produced | generated by projecting 3D map data based on the conditions designated by step S10 (step S14).

Next, the system 100 generates a projection image for processing by projecting 3D map data (step S16). As shown in the figure, the projection conditions for this are the three conditions of hiding the texture of the building, unifying the color of the building, and setting the lighting from directly above to below. . In the present embodiment, the colors of all the buildings are common, but the color to be used may be different for each building.
These conditions are preset in the projection condition designating unit 110. Conditions for projection, that is, conditions such as the line-of-sight direction and the projection method are the same as those for normal projection image generation (step S14).
By doing so, a projected image of a monochromatic building without a pattern or the like can be obtained. Since the obtained projection image for processing is set from directly above to below the lighting, the roof surface of the building is an image with the highest brightness.

The system 100 generates a rooftop image based on the obtained image (step S18). That is, as described above, the processing projection image is an image having high brightness on the roof surface, and therefore, only the portion corresponding to the roof surface is extracted based on the brightness difference.
The state of processing is shown in the figure. The left side is an example of the processing projection image. Two buildings having roof top surfaces S1 and S2 are drawn. Although the color of all surfaces of the building is unified (see step S16), due to the lighting effect, the roof surfaces S1 and S2 have high brightness, and the side surfaces (the hatched portions in the figure) are dark surfaces. It has become.
If only the bright portions S1 and S2 are extracted from these, the state shown on the right side of the figure is obtained. An image in which only the portion corresponding to the rooftop is extracted in this way is called a rooftop image.

Next, the system 100 generates a boundary line based on the rooftop image (step S20). As described later, the boundary line is a line corresponding to the outer periphery of the texture on the roof surface. In this embodiment, the roof surface image is contracted to extract the boundary.
An example of processing is shown in the figure. The left end is the rooftop image. Only one of S1 and S2 previously shown in step S18 of FIG. 2 is shown.
Shrinkage processing is performed on the rooftop image. Shrinkage processing is a well-known image processing technique, and is processing for deleting pixels existing in a predetermined range in the outer peripheral portion of a certain area, such as a rooftop image. How much shrinkage is applied can be arbitrarily set.
The system 100 extracts the boundary based on the rooftop image subjected to the shrinkage process. The frame-like image obtained in this way is the boundary line image generated in step S20.

  The generation of the boundary line image is not limited to the method described above, and can be performed by various methods. For example, a method may be used in which a boundary line is extracted based on a rooftop image, vectorized and recognized as one closed figure, and then subjected to a process of reducing at a certain aspect ratio.

Next, the system 100 generates a rooftop cut image (step S22). The contents of this processing are shown in the figure.
First, as shown in the upper left, the rooftop image is read and subjected to shrinkage processing. This contraction process is the same as that performed in the boundary line generation (step S20). Here, the image obtained by the shrinking process is called a cut frame.
Next, the normal projection image is read as shown in the lower left. And a cut frame is applied to this and only the part which overlaps with a cut frame is extracted as a rooftop cut image. By doing so, as shown in the lower right, a portion of the normal projection image that excludes a constant width of the outer periphery is extracted as the roof top cut image. Since the roof top cut image is a normal projection image cut out, it is not represented by a single color like the roof top image or the cut frame, but is an image reflecting the material of each building.

The system 100 adds a shadow to the generated rooftop cut image and generates a rooftop texture (step S24).
An example of processing is shown in the figure. Shown on the left is the rooftop cut image generated in step S22.
The system 100 shades the two sides of the roof top cut image. In the example in the figure, an example is shown in which shadows are added to the upper and right sides. The shadow widths d1 and d2 can be arbitrarily set. The side to be shaded can also be set arbitrarily, but it is preferable to determine it based on the lighting of the normal projection image. For example, in the normal projection image, when the lighting direction is set from the upper right side to the lower left direction in the drawing, two sides close to the lighting direction, that is, two sides in the drawing are shaded. On the other hand, if lighting is set from the lower left to the upper right in the figure, the left side and the lower side may be shaded.

  The above-described boundary line generation processing (step S20), rooftop cut image generation (step S22), and roof texture generation (step S24) may be switched in order, or both may be performed in parallel. .

The system 100 synthesizes the normal projection image, the boundary line, and the rooftop texture based on the results obtained in the above processes (step S26).
An example of processing is shown in the figure. The left side of the upper stage is a building image obtained by a normal projection image. In this state, the rooftop is depicted as just a flat surface. The boundary line generated in step S20 is shown in the center, and the rooftop texture generated in step S24 is shown on the right side.
The lower part shows the synthesized state. The roof surface is shaded by the roof texture, and a boundary line is drawn on the outer periphery of the roof texture (in the figure, the boundary line is intentionally emphasized). By carrying out like this, the state where the wall of a fixed width is formed in the circumference | surroundings of a rooftop can be represented. The width of the wall is determined by the degree of contraction processing in steps S20 and S22, and the height of the wall is determined by the shadow widths d1 and d2 in step S24.

  The system 100 outputs the image obtained by the synthesis as a 3D map image (step S28). As described above, the 3D map image can be displayed on the smartphone 300 or other display, or can be printed out by the printer 200.

C. Processing example:
5 to 7 are examples of 3D map image data generation processing.
FIG. 5A is a normal projection image of the drawing range designated by the operator (see step S14 in FIG. 2). An example of parallel projection is shown. In this state, each building is drawn with a flat surface on the roof. Hereinafter, a processing example will be described while paying attention to the lower right building BLD. This building BLD also has a flat roof surface SR. In addition, the building BLD is not a simple rectangular parallelepiped but has a feature that there is a recess as shown in the region a.

  FIG. 5B is an example of the processing projection image (see step S16 in FIG. 2). This shows the result of projecting with the lighting from directly above, unifying the color of the entire building. By performing lighting from directly above in this way, the roof is drawn with a higher brightness than the side. The brightness of the rooftop PR is also high for the building BLD. Further, the recess b can be clearly distinguished.

  FIG. 6A is an example of a rooftop image (see step S18 in FIG. 2). The result of extracting only a portion with high brightness from the processing image of FIG. It can be seen that the roof surface PR of the building BLD is extracted in a state where the recess c exists.

  FIG. 6B is an example of a boundary line (see step S20 in FIG. 3). It is the result of extracting the outer periphery as a boundary after shrinking the rooftop image of FIG. A boundary line LR where the concave portion d exists is extracted from the roof surface PR of the building BLD.

  Fig.7 (a) is an example of a roof surface texture (refer FIG.4 S24). An image obtained by cutting out the roof portion of the normal projection image (FIG. 5A) with the roof top cut image generated based on the roof image (FIG. 6A), and then shading the upper side and the right side. It is. The roof texture TR of the building BLD is also generated as shown.

FIG. 7B is a 3D map image. As shown for the lower right building BLD, the normal projection image SR (FIG. 5A) is combined with the boundary line LR (FIG. 6B) and the roof texture TR (FIG. 7A). This is a generated image.
In the 3D map image, a feeling of unevenness is expressed on the roof surface of each building so that a wall with a predetermined width is formed along the outer periphery, and a normal projection image (FIG. 5) is shown as a flat surface. Reality is improved over (a)).

According to the 3D map image data generation system of the present embodiment described above, the reality of the three-dimensional map can be improved by giving the rooftop of the building a feeling of unevenness. Since the unevenness can be automatically given by the processing shown in FIGS. 2 to 4, it does not put a great load on the operator.
Although the unevenness given in the present embodiment does not necessarily reproduce the actual state of the rooftop of the building, as shown in FIG. 7B, even if the reality is not faithfully reproduced, 3 The reality of a three-dimensional map can be improved and the usefulness as a map can be improved.

The image processing for imparting the unevenness shown in the present embodiment is merely an example, and various modifications can be configured as described below.
For example, in addition to the processing shown in the present embodiment, pre-processing or post-processing by an operator may be performed to further improve the reality.
Moreover, although the present Example showed the example of a process which attaches a wall-like uneven feeling along the outer periphery of a rooftop, it represents various equipment bodies, such as a drainage ditch for distributing rain water on a rooftop, and a tank for waterworks. You may give the process which gives an unevenness | corrugation feeling.
Furthermore, it is possible to prepare a plurality of types of processing for giving a feeling of unevenness on the roof and to use them according to the area and shape of the roof.
In the present embodiment, an example of parallel projection processing is shown, but the present invention can also be applied to perspective projection.

  INDUSTRIAL APPLICABILITY The present invention can be used to generate data for drawing a three-dimensional map that represents a feature three-dimensionally as data that can be used for display on a display, print output, and the like.

DESCRIPTION OF SYMBOLS 100 ... Three-dimensional map image data generation system 101 ... 3D map database 102 ... Normal projection image 103 ... Roof top image 104 ... Border image 105 ... Roof texture 106 ... 3D map image 110 ... Projection condition designation | designated part 111 ... Projection process part 112 ... Boundary line generation unit 113 ... Roof texture generation unit 114 ... Image composition unit 200 ... Printer 201 ... Map 300 ... Smartphone

Claims (6)

A 3D map image data generation system for generating 3D map image data for rendering a 3D map representing a building in 3D,
A 3D map database storage unit storing a 3D model representing the 3D shape of the building;
A projection processing unit that projects the three-dimensional model by a predetermined projection method to generate a normal projection image of a three-dimensional map;
Using the image of the area corresponding to the rooftop of the building among the normal projection images generated by the projection processing unit, the rooftop texture of each building is subjected to image processing for giving a sense of unevenness based on a predetermined rule. A roof texture generator for generating
A three-dimensional map image data generation system comprising: an image composition unit that combines the roof texture with the normal projection image to generate the three-dimensional map image data. The three-dimensional map image data generation system according to claim 1,
The roof texture generation unit generates a roof top cut image obtained by cutting an edge of a region corresponding to the roof top, and generates the roof texture by a process of applying a shadow to the roof top cut image. Data generation system. The three-dimensional map image data generation system according to claim 1 or 2,
Furthermore, it has a boundary line generation unit that generates a boundary line image for attaching to the roof top surface by forming a contour line of a region cut out of the edge corresponding to the roof top surface,
The image composition unit is a three-dimensional map image data generation system that further composes the boundary line image. A three-dimensional map image data generation system according to any one of claims 1 to 3,
The projection processing unit unifies the colors of the surfaces constituting each building in the three-dimensional model, generates the processing image by performing the projection in a lighting state from directly above, and in the processing image, A rooftop image is generated by extracting the area represented by the brightness higher than the surface as the rooftop,
The rooftop texture generation unit is a three-dimensional map image data generation system that uses the rooftop surface image to specify a region to be subjected to the image processing in the normal projection image. A three-dimensional map image data generation method for generating three-dimensional map image data for drawing a three-dimensional map representing a building three-dimensionally by a computer,
Reading the 3D model from a 3D map database storage unit storing a 3D model representing the 3D shape of the building;
Projecting the three-dimensional model by a predetermined projection method to generate a normal projection image of a three-dimensional map;
A step of generating a roof texture of each building by performing image processing for giving an uneven feeling based on a predetermined rule using an image of an area corresponding to the roof surface of the building in the generated normal projection image When,
Combining the roof texture with the normal projection image to generate the three-dimensional map image data. A computer program for generating three-dimensional map image data for drawing a three-dimensional map representing a building three-dimensionally by a computer,
A function of reading the 3D model from a 3D map database storage unit storing a 3D model representing the 3D shape of the building;
A function of projecting the three-dimensional model by a predetermined projection method to generate a normal projection image of a three-dimensional map;
A function of generating a roof texture of each building by performing image processing for giving an uneven feeling based on a predetermined rule using an image of an area corresponding to the roof surface of the building among the generated normal projection images When,
A computer program for causing a computer to realize a function of synthesizing the rooftop texture with the normal projection image and generating the three-dimensional map image data.