JPH08272851A - Method and device for preparing grained pattern having cross-section - Google Patents

Abstract

PURPOSE: To artificially generate a natural grained pattern close to that of natural tree or the grained pattern improved in design which is not obtained in natural tree. CONSTITUTION: A growth ring model Mr composed of the set of picture elements provided with a picture element value cyclically changed based on the distance from the center axis C is prepared, vessels F extended in a direction approximately along the center axis C are arranged inside the growth ring model Mr, a vessel model Mf composed of the set of the picture elements present at least in a contour part inside the conductor F is added to the growth ring model Mr and thus, a three-dimensional tree model M is prepared. At the time of cutting the three-dimensional tree model M by a prescribed cutting surface J, a pattern constituted of the set of the picture elements positioned on the cutting surface J is extracted as the grain pattern. As the cutting surface J, the wrinkled cutting surface having fluctuation by a two-dimensional fractal field is used.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for producing a wood grain pattern having a conduit cross section, and in particular, a wood grain pattern having a natural texture or a wood grain pattern having a high design property is artificially formed. The technology for generating.

[0002]

2. Description of the Related Art A wood grain pattern is widely used as a building material product such as wallpaper and a package pattern for various products. When a printed matter having such a wood grain pattern is created, a method is usually used in which the grain of natural wood is photographed with a camera and the grain pattern of the natural wood is used as it is. Further, in recent years, image processing technology using a computer has become widespread also in the printing field. Therefore, a grain pattern pattern of a natural tree is captured as image data by a CCD camera or the like, and a computer is used for this image data. And perform the necessary image processing,
A method of printing based on the processed image data is also widely used.

Generally, the wood grain pattern includes an annual ring pattern and a conduit cross-section pattern. The annual ring pattern is
It is a pattern that is formed as the tree grows year by year. Usually, a difference in light and shade occurs based on the difference in temperature and temperature in the growth environment of trees, and this difference in light and shade appears as an annual ring pattern. Therefore, it becomes a periodic light and shade pattern every year. On the other hand, the conduit cross-sectional pattern is a cross-sectional pattern obtained by cutting the conduit of the tree. The conduit is an organ necessary for a tree to perform a physiological function as a plant, and is a thin tube extending from the trunk toward the treetop, and its cross section is generally elongated and elliptical. Therefore, when observing the grain pattern that appears in the grain of natural wood, the annual ring pattern is generally recognized, but when viewed in detail, it is recognized that many small conduit cross-section patterns are arranged.

In a wallpaper or the like, it is common to express a wood grain pattern by superimposing an annual ring pattern and a conduit cross-section pattern in order to faithfully reproduce the texture of the above-mentioned wood grain pattern of natural wood. is there. Usually, a method is adopted in which the annual ring pattern and the conduit cross-section pattern are separately photographed from the grain of natural wood, separate plates are created, and both are combined at the time of printing. The annual ring pattern and the conduit cross-sectional pattern may both be formed on a medium such as a vinyl chloride sheet by printing, or the annual ring pattern may be printed and the conduit cross-sectional pattern may be formed separately by the embossed concavo-convex structure. There is also. Originally, since the conduit cross section of natural wood has an uneven structure, if the conduit cross section pattern is formed as an embossed uneven structure, a texture closer to that of natural wood can be expressed.

[0005]

However, the work of extracting the annual ring pattern and the conduit cross-section pattern from the grain of natural wood is a very troublesome work. In order to extract a pattern with high designability, it is necessary to start by selecting a plate material having an annual ring pattern and a conduit cross-sectional pattern that are excellent in design. Moreover, it is necessary to prepare a plate material in which a pattern clearly appears so that a clear pattern can be extracted by photographing. Particularly, the work of extracting a clear conduit cross-section pattern from the grain of natural wood is technically extremely difficult. This work is usually performed by taking a photograph of the cross section of the sheet with a camera and capturing the conduit cross section pattern as digital data from the photograph with a scanner device, or by capturing the conduit cross section pattern directly from the sheet with a digital camera. . However,
The spatial resolution of cameras and scanners is limited, and it is difficult to faithfully capture the shape of a minute conduit cross section as pattern data. In particular, depending on the natural wood, there may be a slight difference in color tone between the conduit portion and the non-conduit portion appearing in the wood grain, and for such wood grain, the sensitivity of the image input system, the dynamic range, Due to the limit of the number of quantization bits, the number of A / D bits, etc., it is very difficult to faithfully capture the conduit cross-section pattern. In such a case, a method is used in which only the conduit section is colored with a dye and then a photo is taken, but it is technically difficult to accurately color only the conduit section. Therefore, there is a problem that the time and effort are increased.

Further, the wood grain pattern pattern extracted from the natural wood by the conventional method has a problem that the design is poor. That is, if the annual ring pattern and the conduit cross-section pattern are extracted from the grain of natural wood, it is certainly possible to reproduce a pattern with a natural texture close to that of natural wood on the medium, but conversely, Since it is limited to the woodgrain pattern found in naturally existing trees, it becomes impossible to create a woodgrain pattern having a novel conduit design with a conduit cross section. As a result, the pattern tends to be monotonous in terms of design.

Therefore, the present invention provides a method and an apparatus for creating a wood grain pattern which can artificially generate a natural wood grain pattern pattern close to that of natural wood, or a wood grain pattern pattern of high design which is not found in natural wood. The purpose is to provide.

[0008]

[Means for Solving the Problems]

(1) A first aspect of the present invention is a method of artificially creating a woodgrain pattern having a conduit cross section, which is a set of pixels having pixel values that periodically change based on a distance from a predetermined central axis. And a conduit model consisting of a set of pixels existing at least in the contour part of the conduit is arranged in the annual ring model. The step of creating a three-dimensional tree model by adding it to the model, and when this three-dimensional tree model is cut by a predetermined cutting plane, a pattern composed of a set of pixels located on the cutting plane The extraction step is performed as follows.

(2) A second aspect of the present invention is the above-mentioned first aspect.
In the method of creating a woodgrain pattern according to the aspect of (1), a pixel value within a predetermined range is defined for each pixel existing in the area outside the conduit, and each pixel existing in the area inside the conduit is defined by the above predetermined value. A temporary pixel value outside the range is defined, and the temporary pixel value is replaced with a pixel value based on the depth from the cut surface to the contour of the conduit.

(3) A third aspect of the present invention is the above-mentioned first aspect.
In the method of creating a wood grain pattern according to the aspect of (3), the three-dimensional shape of the conduit groove formed by cutting the three-dimensional tree model is recognized, and predetermined light source conditions are set to reflect light from each part of the conduit groove. The intensity is calculated and the pixel value of each pixel existing in the area inside the conduit tube is determined by the calculated reflected light intensity.

(4) A fourth aspect of the present invention relates to the above-mentioned first aspect.
In the method for creating a woodgrain pattern according to the third aspect, the arrangement density of the conduits is determined based on the distribution of pixel values of the annual ring model.

(5) A fifth aspect of the present invention relates to the above-mentioned first aspect.
-In the method for creating a woodgrain pattern according to the fourth aspect, a two-dimensional fractal grid in which a predetermined scalar value is self-similarly defined at each grid point on a two-dimensional plane is prepared, and each of the two-dimensional fractal grids is prepared. A wrinkle cut surface is generated by displacing each point on the reference plane in a predetermined direction according to the scalar value of the grid point, and the three-dimensional tree model is cut by this wrinkle cut surface to form a grain pattern. It is designed to be extracted.

(6) A sixth aspect of the present invention relates to the above-mentioned first aspect.
~ In the method of creating a woodgrain pattern according to the fifth aspect, a predetermined scalar value is defined in a two-dimensional fractal grid or three-dimensional space defined in a self-similar manner to each grid point on a two-dimensional plane. The 3D fractal grid is prepared, and each pixel that makes up the 3D tree model is displaced based on the scalar value defined at each grid point of the prepared 2D or 3D fractal grid. A three-dimensional tree model is created, and the distorted three-dimensional tree model is cut along a predetermined cutting plane to extract a grain pattern.

(7) A seventh aspect of the present invention is, in an apparatus for artificially creating a wood grain pattern having a conduit cross section, having a pixel value that periodically changes based on a distance from a predetermined central axis. An annual ring model generating means for generating an annual ring model composed of a set of pixels, and a conduit extending in a direction substantially along the central axis in the annual ring model are arranged, and from the set of pixels existing in at least the contour portion in the conduit. A three-dimensional tree model generating means for generating a three-dimensional tree model by adding the above conduit model to the annual ring model, and a two-dimensional fractal in which a predetermined scalar value is self-reciprocally defined at each grid point on the two-dimensional plane. A two-dimensional fractal grid generating means for generating a grid and a three-dimensional tree model are installed by a predetermined wrinkled cut surface having fluctuations according to the scalar value of the two-dimensional fractal grid. The pattern formed by the set of pixels located on this wrinkle cut surface,
And a wood grain pattern pattern extracting means for extracting as a wood grain pattern pattern.

(8) An eighth aspect of the present invention is the seventh aspect described above.
In the device for creating a woodgrain pattern according to this aspect, a random number generating means for generating a random number within a predetermined numerical range is further provided, and the two-dimensional fractal grid generating means uses the random number generated by the random number generating means to generate a two-dimensional pattern. The fractal grid is generated, the annual ring model generation means defines the length of each cycle using the random numbers generated by the random number generation means, and the three-dimensional tree model generation means generates the random number generation means. The arrangement of conduits is performed using random numbers.

(9) A ninth aspect of the present invention is based on the above-mentioned seventh aspect.
Alternatively, in the grain pattern creating apparatus according to the eighth aspect, a color map setting unit that sets a color map for making the pixel value of each pixel correspond to a plurality of primary color components necessary for printing, and a color map setting unit. Based on the color tone allocation means for allocating a plurality of primary color components necessary for printing to each pixel forming the wood grain pattern pattern extracted by the wood grain pattern pattern extraction means, and the primary color components allocated by this color tone allocation means, And a printing unit that prints a grain pattern by using the printing unit.

[0017]

[Operation] According to the method and apparatus for producing a woodgrain pattern having a conduit cross section according to the present invention, it is possible to artificially generate a woodgrain pattern using a computer. In the present invention, first, on a computer,
A three-dimensional tree model is built. This three-dimensional tree model is a tree ring model with a conduit model added, and is a model of the actual three-dimensional form of a natural tree. That is, this three-dimensional tree model is configured by arranging pixels having a predetermined pixel value in the three-dimensional space. Here, the annual ring model is composed of a set of pixels having a pixel value that periodically changes based on a distance from a predetermined center axis, and reproduces a light and shade pattern that appears based on annual growth of trees. Become a model. Also, the conduit model is
It is a model showing the conduits placed inside the annual ring model. If such a three-dimensional tree model is cut at a predetermined cutting plane and a pattern is formed by a set of pixels located on this cutting plane, the pattern will be the grain that appears on the cutting plane when the natural tree is cut. It corresponds to the pattern.

If the grain pattern is created by such a method, the work of extracting the annual ring pattern and the conduit cross-section pattern from the actual natural tree is not necessary at all, and it is possible to artificially create the grain pattern. Become. Moreover, if the three-dimensional tree model is close to an actual natural tree, a natural grain pattern close to that of a natural tree can be obtained. In addition, if the parameters for constructing the three-dimensional tree model are variously changed, it is possible to obtain a wood grain pattern pattern with a high design which is not found in natural trees.

With respect to the conduit cross-section pattern obtained by cutting the conduit model, the depth from the cut surface to the conduit wall can be obtained by calculation. Therefore, by replacing this depth information with pixel values, It becomes possible to give more accurate pixel values for the pixels inside the cross-sectional pattern.

If the three-dimensional shape of the conduit groove formed by cutting the three-dimensional tree model is recognized, shadow information can be added to the inner wall of the conduit groove by using a so-called rendering method. With respect to the pixels inside the conduit cross-section pattern, if the pixel value expressing this shadow is given, the stereoscopic effect of the conduit groove can be expressed even if the conduit cross-section pattern is printed on a flat surface.

Generally, the annual ring pattern has a gradation of light and shade in an annual cycle, but similarly, the distribution of the conduit cross-section pattern also changes in an annual cycle. Usually, the lighter the density of the annual ring pattern, the higher the conduit density, and the darker the density, the lower the density.
This is because the growth pattern of trees differs depending on the growing season (four seasons), and the density of the generated conduits also varies depending on the growing season (four seasons). Therefore, when creating the conduit model, the density of the conduit cross-section pattern in the obtained woodgrain pattern can be determined by determining the density of the conduit pattern based on the distribution of pixel values of the annual ring model. It will be in tune with the distribution, and the overall natural texture can be expressed despite the artificial generation of the conduit cross-section pattern.

Any surface may be used as a cutting surface for the three-dimensional tree model, but in order to obtain a wood grain pattern having a natural texture, a surface having fractal fluctuation is used as a cutting surface. Is preferred. For example, by using a two-dimensional fractal grid in which a given scalar value is self-similarly defined for each grid point on the two-dimensional plane, by displacing each point on the plane according to the scalar value of each grid point, If a wrinkled surface is created, a three-dimensional tree model is cut by this wrinkled surface, and this wrinkled cutting surface is projected again onto a two-dimensional plane, the resulting grain pattern will potentially contain fractal fluctuations. This is a pattern that reflects the fluctuations of nature. Further, instead of giving fractal fluctuation to the cut surface, it is possible to give fractal fluctuation to the three-dimensional tree model itself. For example,
Using a two-dimensional or three-dimensional fractal grid, displace each pixel that makes up the three-dimensional tree model to create a distorted three-dimensional tree model, and cut this with a predetermined cutting plane to obtain a grain pattern. Good.

Random numbers are preferably used in the process of generating the annual ring model, the conduit model and the fractal grid. The pixel value of the pixels that make up the annual ring model changes periodically based on the distance from the central axis.
This cycle is more effective in expressing the natural texture than in a fixed cycle. Therefore, it is preferable to determine this cycle using a random number.
For the conduit model, a natural conduit distribution can be realized if the arrangement of the conduits is determined using random numbers. Furthermore, if a fractal lattice is generated using random numbers, a fractal field having a scalar value within a predetermined range can be easily obtained.

When finally printing the obtained grain pattern on the medium, a color map is prepared for making the pixel value of each pixel correspond to a plurality of primary color components necessary for printing. It is convenient to leave it. If you want to finely adjust the color tone of printing, all you have to do is adjust this color map, and it is possible to express a wood grain pattern with a desired hue on the medium with a relatively simple process. .

[0025]

The present invention will be described below based on illustrated embodiments.

§1. Structure of Wood Pattern Pattern First, the structure of a general wood pattern pattern will be described with reference to FIG. A general wood grain pattern P12 used for decoration such as wallpaper or building material is obtained by superimposing an annual ring pattern P1 and a conduit cross-section pattern P2 as shown in FIG.

As described above, the annual ring pattern is a pattern formed according to the annual growth of trees,
Usually, a difference in light and shade occurs based on the difference in temperature and temperature in the growth environment of trees, and this difference in light and shade appears as an annual ring pattern. The diagram shown in the upper left of FIG. 1 is a simple geometric model diagram showing the natural tree T1, and shows the natural tree T1 as a simple coaxial cylinder model. The area between one cylinder and the outer cylinder shows the portion that has grown over the course of one year. Of course, an actual natural tree does not have such a perfect coaxial cylindrical model, but has a considerably distorted shape. When such a natural tree T1 is cut along a predetermined cut surface J1, the annual ring pattern P1 as shown is obtained on the cut surface. In the case of this simple geometric model, the annual ring pattern P1 becomes a concentric ellipse pattern, and the region between one ellipse and the ellipse outside the one ellipse corresponds to the portion grown in one year.

On the other hand, the conduit cross-section pattern is a pattern obtained by cutting a conduit required for a tree to perform a physiological function as a plant, and is usually a fine, elongated elliptical pattern. The diagram shown in the upper right of FIG. 1 is a simple geometric model diagram showing one conduit T2 that constitutes the tissue of a natural tree, and the conduit T2 is shown as a simple cylindrical model. The substances necessary for the physiological action of the plant are transported along the inside of this cylindrical conduit. When such a conduit T2 is cut by a predetermined cutting plane J2, a single elliptical conduit cross-section pattern P as shown by hatching in the drawing is obtained on the cutting plane.
Of course, in practice, this single conduit cross-section pattern P
Will not be a geometrically perfect ellipse, but rather a rather distorted shape. In the upper right of FIG. 1, one conduit T2 is provided with a cut surface J2.
Although there are many such conduits T2 in the tree, the conduit cross-section pattern P shown in FIG.
As in 2, a pattern in which a large number of conduit cross sections are arranged will be obtained. Usually, when cutting a natural plate, it is often cut in a cross section along the growth direction in order to obtain as beautiful a grain as possible. Therefore, the conduit cross-section pattern P2 is a collection of very elongated and distorted elliptical patterns, and when observed with the naked eye, many linear marks are all arranged substantially in the direction of tree growth. Looks like a simple pattern.

By overlaying such a conduit cross-section pattern P2 on the annual ring pattern P1, a wood grain pattern P1 is obtained.
2 will be obtained. Such a wood grain pattern P
12 is represented on the medium as a ring pattern P.
1 is generally expressed by printing, but a typical method is to express the conduit cross-section pattern P2 by printing and express it as an embossed concavo-convex structure (the inside of the conduit is a recess). . In addition, printing may be performed using a raised ink for the purpose of design effect.

In the case of decorating a printed matter such as wallpaper with a wood grain pattern, conventionally, a tree ring pattern P1 is formed from natural wood.
And the conduit cross-section pattern P2 have been extracted, but this work is extremely labor-intensive, and it has already been mentioned that there is a problem that it is not possible to obtain a woodgrain pattern with a novel design. That's right. In order to solve such a problem, the present invention provides a novel technique for artificially generating a grain pattern P12 using a computer. Examples of this technique will be described below.

§2. Construction of annual ring model First, an annual ring model Mr is constructed on a computer. The annual ring model Mr is a geometrical virtual model composed of a set of pixels having a pixel value that periodically changes based on the distance from a predetermined central axis C, and is an element of the annual ring of the natural tree T1 shown in FIG. Is to express. Figure 2
[Fig. 3] is a perspective view showing a specific example of such an annual ring model Mr. In the present embodiment, such an annual ring model Mr has an annual growth parameter of 1 year average growth width: D (unit mm) and maximum annual growth width variation: Δ (unit mm). It is generated by setting. The annual ring model Mr shown in FIG. 2 is a model in which year = 4 is set, and the growth width D1 in the first year, the growth width D2 in the second year, the growth width D3 in the third year, the growth width D4 in the fourth year. It is shown. Here, the growth width Di in the i-th year is given by the equation: Di = D + ΔRND (1). Note that RND is a random number generated by a random number generating means using a computer, and −1 ≦ R
It takes a range of ND ≦ + 1.

The actual state of the annual ring model Mr is a three-dimensional stereoscopic image, which is a set of pixels distributed in a three-dimensional space.
Therefore, a predetermined pixel value is defined for each pixel. The feature of this model is that a pixel value that changes periodically is given based on the distance from the central axis C. In this embodiment, the pixel value of one pixel is represented by 7-bit data. Therefore, each pixel is given one of 128 pixel values of 0 to 127. In the case of this embodiment, the pixel value given to each pixel at this stage is not a value that is directly used as a gradation value at the stage of printing (as will be described later, by assigning a tone based on a color map, Gradation values used directly in the printing stage are assigned). Therefore, in this specification, 7 defined for each pixel at this stage is used.
The pixel value of the bit will be specifically referred to as “potential value U”.

FIG. 3 is a diagram showing a potential value U defined in each pixel of the annual ring model Mr on a plane perpendicular to the central axis C. In the section of the growth width D1 of the first year, the potential value U is 0 to 12 in the growth direction.
It increases monotonically with 7, but when it enters into the section of the growth width D2 in the second year, the potential value U returns to 0 again, and again 0 to 1
It increases monotonically with 27. In this way, the potential value U monotonically increases from 0 to 127 in the growth width section of each year. FIG. 4 shows such a potential value U
Is a graph showing the periodic change of the. In this embodiment, the monotonic increase of the potential value U is linearly increased, but it is not always necessary to increase linearly. The meaning of "a pixel value that changes periodically based on the distance from the central axis C", which is a feature of this model, is thus described for each growth width section of each year. Potential value U
Means that is given. In the present embodiment, the definition of such a potential value U is 360 around the central axis C.
The same applies to any growth direction of °, and the same applies to any plane perpendicular to the central axis C. Of course, different potential values may be defined for each growth direction of 360 ° or for each surface perpendicular to the central axis C.

After all, if such a three-dimensional image of the annual ring model Mr is constructed on the computer, the potential value U for the pixel existing at an arbitrary position in this three-dimensional space.
Can be obtained by a very simple operation. However, if you consider the actual calculation on the computer,
This annual ring model Mr is not a stereoscopic image composed of a set of pixels having a finite size, but rather a specific potential value at an arbitrary coordinate position (geometric point with volume 0) in a three-dimensional space. U can be regarded as a defined three-dimensional potential field, and in an actual computer, it is not always necessary to store the pixel value of each pixel in the memory, and it is possible to define U as a function. Good.

The individual pixels constituting the annual ring model Mr have the potential value U of 0 to 127 as the pixel value, but when actually printing each pixel on the medium, this The potential value U is converted into the predetermined gradation value C. Therefore, a color map for making the potential value U correspond to the gradation value C is prepared. For example, in the graph shown in FIG. 5, the color map C (U) is shown as a function of the potential value U and the gradation value C. By using this color map C (U), an arbitrary potential value U from 0 to 127 can be converted into any gradation value C within the range from the maximum gradation Cmax to the minimum gradation Cmin. . At the right end of the graph of FIG. 5, the gradation distribution corresponding to the gradation change from Cmax to Cmin is illustrated. On an actual medium, each pixel is represented as a minute area having such a shade.

When a tree ring pattern that is closer to that of a natural tree is expressed on the medium, the color map C (U) is set in consideration of the shading of individual parts at each actual growing time of the natural tree. do it. For example, in the example shown in FIG. 5, the numerical range of 0 to 127 taken by the potential value U is
Growth period over a year, from January 1 to December 3
The gradation value C is determined by associating with one day (in the case of a general tree, the growth rate is different depending on each period, so that it is not a linear correspondence). Specifically, for example, regarding the actual tree, it is confirmed what gradation value C the part that has grown during the period of July 1 (summer) has, and it corresponds to the growth position of July 1. For the potential value U = 95, a color map C (U) may be set so as to correspond to the actual gradation value C of the portion grown during the period of July 1. Generally, in a general tree, a difference in growth based on the difference in temperature between one year and the other causes a difference in light and shade based on the growth period. Therefore, it is enough for the color map C (U) to define the correspondence for one year. is there. Although an example of the color map C (U) for associating the monochrome gradation value C is shown here, when color printing is finally performed, for each pixel, four primary colors such as YMCK are used. A color map C (U) capable of defining the gradation value of will be prepared.

Since such a color map C (U) differs depending on the natural tree to be expressed, a different color map C (U) is created for each natural tree to be imitated.
It is preferable to prepare. In the case of a general tree, since the growth is active from spring to summer and the growth is slow from autumn to winter, the color map C (U) shown in FIG. 5 is used.
As shown in, the part grown from spring to summer becomes a light color, and the part grown from autumn to winter becomes a dark color. However, for the purpose of obtaining a wood grain pattern with a novel design that is not found in natural trees, there is no need to stick to the nature of such natural trees, and a free color map C
You may set (U).

By the way, after all, an annual ring model Mr as shown in FIG. 2 (a model in which pixels having a potential value U of 0 to 127 are arranged in a three-dimensional space as pixel values) is constructed, and this annual ring model Mr is prepared. Is cut by a predetermined cut plane J1, and a color map C (U) as shown in FIG. 5 is applied to a pattern formed of a set of pixels located on the cut plane to gradation the potential value U of each pixel. When converted to the value C, the annual ring pattern P1 as shown in FIG. 1 is obtained. In other words, the natural tree T shown in the upper left of FIG.
1 is actually cut by the cutting plane J1 and the operation of obtaining the annual ring pattern P1 on this actual cutting plane is simulated on a computer. However, the pattern obtained by virtually cutting only such annual ring model Mr is only the annual ring pattern P1. The actual grain pattern P12 is, as shown in FIG.
1 to which the conduit cross-section pattern P2 is added. Therefore, a method for adding the conduit cross-section pattern P2 will be described below.

§3. Addition of conduit model / 3D tree model
Construction of Dell The three-dimensional tree model M in the present invention is obtained by adding the conduit model Mf to the annual ring model Mr constructed in §2. FIG. 6 is a perspective view showing a specific example of the three-dimensional tree model M constructed in this way. The conduit model Mf has a conduit F extending in a direction substantially along the central axis C.
Is a model configured by arranging a large number of pixels in the annual ring model Mr, and more specifically, a model configured by a set of pixels existing at least in the contour portion in each individual conduit F. In this embodiment, each of the conduits F is composed of a cylindrical solid body parallel to the central axis C, and the conduit model Mf is composed of pixels existing in the entire space inside the cylindrical solid body. . Therefore, of the space occupied by the annual ring model Mr constructed in §2,
The space occupied by the cylindrical conduit F will be replaced by the conduit model Mf.

FIG. 7 is a diagram showing pixel values defined for each pixel of the three-dimensional tree model M on a plane perpendicular to the central axis C. For the three-dimensional tree model M of FIG. 6, if an XYZ three-dimensional coordinate system is defined as shown,
FIG. 7 corresponds to a cross section of the three-dimensional tree model M cut by a plane parallel to the XY plane. In FIG. 7, the area outside the conduit F is composed of pixels forming the annual ring model Mr, and the area inside the conduit F is composed of pixels forming the conduit model Mf. As described above, the potential value U having a value of 0 to 127 is periodically changed in the pixels forming the annual ring model Mr (pixels outside the conduit F) based on the distance from the central axis C. It is defined as a pixel value. Therefore, for example, the pixel K1 in FIG. 7 has a pixel value of "64." On the other hand, in the pixels (pixels inside the conduit F) forming the conduit model Mf, in this embodiment, all pixels are equal to “12”.
The pixel value "8" is given. For example, the pixel K2 in FIG. 7 has a pixel value of “128”, and all the pixels inside the multiple conduits F have the pixel value of “128”. The pixel value “128” is, so to speak, a temporary pixel value, and is for indicating that the pixel is a pixel existing inside the conduit F. Therefore, the conduit F
Pixels outside the pixel have a pixel value in the range of 0 to 127, but the pixel value given to the pixel inside the conduit F is
Any pixel value may be defined as long as it is a pixel value outside the range of 0 to 127.

After all, the three-dimensional tree model M shown in FIG.
From a set of pixels having a potential value U of 0 to 127 (pixels outside the conduit F) and pixels having a fixed value of “128” as a pixel value (pixels inside the conduit F) It will be a model. In this embodiment, all the pixels located in the space inside the conduit F constitute the conduit model Mf.
Need only be configured with pixels that exist in at least the contour portion (tube wall portion) of the conduit F. In the present invention, in short, it suffices to be able to recognize whether the predetermined point in the three-dimensional space is an internal point or an external point of the conduit F,
Such recognition is possible if at least the contour of the conduit F is defined.

By the way, in order to construct the three-dimensional tree model M having a natural texture as much as possible, the conduit F is arranged at a predetermined position inside the annual ring model Mr so as to satisfy the following two conditions. It should be set to. The first condition is to arrange them at random positions as much as possible, and the second condition is to arrange them so that the arrangement density of the conduits F is determined based on the distribution of the pixel values of the annual ring model Mr. Is. The first condition may not require special explanation. The regular arrangement of the conduits F enhances the artificial texture. In order to express the natural texture peculiar to natural wood, it is necessary to generate the conduits F at random positions.
The second condition is based on the fact that the conduit density in natural wood varies with the time of growth. For example, the density of conduits in the part that grew in January (winter) and July (summer)
There is a big difference between the density of the conduits in the area that grew up to the same area and the same tree, based on the difference in temperature.

FIG. 8 is a graph showing an example of the correspondence relationship between the growth time t and the conduit density D. What kind of graph is specifically obtained depends on the individual tree, but if only a tree can be specified, an average graph for that tree can be obtained by actual measurement. The graph shown in FIG. 8 is an example of a general tree in which the duct density is high in the portion grown from spring to summer and is low in the portion grown from autumn to winter. When a large number of conduits F are generated in the annual ring model Mr to construct the three-dimensional tree model M, if the conduits F are generated with the density D shown in such a graph, it becomes closer to a natural tree. It becomes possible to construct a three-dimensional tree model M having a conduit density distribution.

As described above, the annual ring model Mr constructed
The pixel value (potential value U) of each pixel can be associated with the growth period. Therefore, in the graph shown in FIG. 8, it becomes possible to associate potential value U → growth time t → conduit density D. Therefore, in this embodiment,
By associating this conduit density D with a probability value Z indicating the probability of occurrence of the conduit F, and using this probability value Z, an arrangement that satisfies both the above-mentioned first condition and second condition is provided. In addition, a large number of conduits F are generated. The probability value Z is a value proportional to the conduit density D, that is, Z =
It can be obtained as kD. Here, k is a proportional constant, and X of one pixel forming the constructed annual ring model Mr.
It is a value obtained by dividing the occupied area on the Y plane by the unit area. For example, if the occupied area of one pixel on the XY plane is 1/100 of the unit area, k = 1/100,
Z = 1/100 · D. Specifically, for example, X
If there are three conduits F in the unit area on the Y plane, the conduit density is D = 3, so the probability value Z = 0.03, and the conduits F can be generated at a rate of three per 100 pixels. I understand that it is good.

After all, for each pixel, it becomes possible to associate potential value U → growth time t → conduit density D → probability value Z, and a probability function Z (U) as shown in FIG.
Can be defined. Such a probability function Z (U)
In practice, it suffices to prepare a table for converting an arbitrary potential value U within the range of 0 to 127 into a predetermined probability value Z.

As a concrete process for generating the conduit F, the following process using random numbers is performed in this embodiment. First, with respect to each of the pixels on the XY plane of the annual ring model Mr, it is determined whether or not the “occurrence position” of the conduit F is reached as follows. That is, for the potential value U (pixel value) of the pixel, the probability function Z (U) shown in FIG. 8 is applied to obtain the corresponding probability value Z (actually, the probability value Z is obtained by the operation of drawing a table. obtain).
Subsequently, an arbitrary random number RND (0 <RND <1) is generated, and if RND <Z, the pixel is set as the “generation position”, and if RND ≧ Z, it is not set as the “generation position”. If such a process is performed separately for all the pixels on the XY plane, it can be finally determined whether or not the "occurrence position" is reached for each pixel.

In this way, when the pixel to be the "occurrence position" can be determined, an elongated cylindrical solid body having a center line which is a line passing through the center of the pixel and parallel to the Z axis is generated, and this is formed into one conduit F Can be defined as In this way, a large number of conduits F passing through the pixels at the "occurrence position" are generated as shown in FIG. 6, and the distribution density thereof is based on the potential value U of the pixels of the annual ring model Mr. , And the placement of the individual conduits F will be random. In other words, a large number of conduits F can be generated in an arrangement that satisfies both the first condition and the second condition described above.

The large number of conduits F to be generated preferably have random diameters. In this embodiment, an average diameter φ and a variance σ ^{2 of} the diameter are predetermined, and a random number RND is used to obtain the average diameter φ and the variance σ ^2. Conduit F
I am trying to generate. After all, in this embodiment, as shown in the right side of FIG. 6, the probability function Z (U) is used as the conduit parameter, and the actual size length L of one pixel (the occupied area of one pixel on the XY plane is L ²⁾ . ), The average diameter φ of the conduit and the variance σ ² of the diameter are determined in advance, and using these conduit parameters, a large number of conduits F are generated in the annual ring model Mr to construct the three-dimensional tree model M. . Any conduit Fi thus generated has a diameter φi and X
It can be expressed by three variables of position coordinate values (xi, yi) on the Y plane. Alternatively, the position coordinate value (x
Polar coordinate values (ri, θi) may be used instead of i, yi).

§4. Cutting of Three-Dimensional Tree Model Now, if the three-dimensional tree model M can be constructed in this way, a grain pattern can be obtained by cutting the model M at a predetermined cutting plane J as shown in FIG.
In other words, the pattern composed of a set of pixels located on the cut surface J is a wood grain pattern. Depending on the arrangement angle of the cutting plane J, the pattern of the obtained grain pattern will be slightly different, but this point is exactly the same as the grain pattern obtained by cutting the natural wood.

FIG. 10 is a diagram showing an example of the grain pattern P12 obtained in this way. Since the three-dimensional tree model M is composed of the annual ring model Mr and the conduit model Mf, the obtained wood grain pattern P12 also
The annual ring pattern P1 and the conduit cross-section pattern P2 are included. In this embodiment, the pixel K3 forming the annual ring pattern P1 has pixel values 0 to 127 (potential value U).
And the pixel K4 forming the conduit cross-section pattern P2 is
It will have a temporary pixel value of 128. Therefore, 12
The area having a pixel value of less than 8 is the annual ring pattern P1, and the area having a pixel value of 128 is the conduit cross-section pattern P.
You can immediately recognize that it is 2.

To print such a woodgrain pattern P12 on the medium, for pixels having pixel values 0 to 127, a color map C (U) as shown in FIG. 5 is used.
Each potential value U may be converted into the gradation value C. Thereby, the annual ring pattern P1 can be expressed on the medium. In the case of color printing, instead of the monochrome gradation value C, a color value that is a combination of the primary color components of YMCK is used. On the other hand, with respect to the pixel having the provisional pixel value 128, that is, the pixel forming the conduit cross-section pattern P2, some handling methods can be considered. The simplest handling method is to uniformly assign some suitable gradation value to represent the conduit groove. For example, by assigning a brown color to the pixel values 0 to 127 and assigning a black color to the pixel value 128, the black conduit cross-section pattern P is included in the brown annual ring pattern P1.
It is possible to express a wood grain pattern P12 in which two are mixed. However, with this method, the inside of the conduit cross-section pattern P2 has a uniform color, and therefore the texture of natural wood cannot be expressed in detail. Therefore, in this embodiment, the pixel value distribution is defined in the conduit cross-section pattern P2 by the following method.

Now, as shown in FIG. 11, one conduit F
Consider a model in which is cut at the cutting plane J. In this model, if the portion below the cut surface J constitutes a plate material, a conduit groove G as shown in the figure will be formed in the surface layer of the plate material. The contour line when the conduit groove G is viewed from above is nothing but a single conduit cross-sectional pattern P. If the conduit F is a perfect cylindrical figure and the cut surface J is a plane, this single conduit cross-sectional pattern P is a geometrically perfect elliptical pattern. Here, the contour of the conduit F (tube wall)
Considering the depth distribution in the conduit groove G by defining the distance between each point above and the cut surface J as the depth of the conduit groove G, as shown in the drawing,
It can be seen that along the major axis of the ellipse, the depth increases monotonically from one side to the other side, from the shallow portion to the deep portion. Specifically, the depth Z0 at the right end position of the single conduit cross-sectional pattern P shown in FIG. 11 is the shallowest, and the depth Z at the position C1.
1, the depth Z2 at the position C2, and the depth Z3 at the position C3 are gradually increased in this order, and the depth Z4 at the left end position of the pattern P is the deepest. Such a depth distribution is in the major axis direction of the ellipse, and another depth distribution is obtained in the minor axis direction of the ellipse. In any case, the depth distribution at each position inside the single conduit cross-section pattern P is unique if two parameters of the diameter of the conduit F and the angle between the conduit F and the cut surface J are known. Set to. Therefore, the three-dimensional tree model M as shown in FIG.
Assuming that a wood grain pattern P12 as shown in FIG. 10 is obtained by defining the cut surface J and the cut surface J, predetermined positions are set for respective positions inside the individual conduit cross-section patterns P in the wood grain pattern P12. The depth can be calculated.

By the way, as shown in FIG. 10, at the present stage, a temporary pixel value "128" is uniformly given to the pixels K4 constituting the conduit cross-section pattern P2.
Therefore, in this embodiment, a process of replacing the temporary pixel value with the original pixel value of 128 to 255 based on the depth is performed. For example, if the pixel value 128 is associated with the depth Z0 shown in FIG. 11 and the pixel value 255 is associated with the depth Z4, then 128 to 255 for the intermediate depths Z1, Z2, and Z3. A predetermined depth within the range is associated with, and for all pixels within a single conduit cross-section pattern P, any pixel value from 128 to 255 will be associated. After all, in this embodiment, finally, the pixel K3 forming the annual ring pattern P1 is 0.
Pixel values in the range of to 127 are assigned, and 128 to 255 are assigned to the pixels K4 constituting the conduit cross-section pattern P2.
Pixel values in the range will be assigned. And
Similarly to the case where the color map C (U) as shown in FIG. 5 is prepared for the pixel values 0 to 127, the pixel values 128 to 25
If a different color map for expressing the inside of the conduit groove G is prepared for 5 as well, it is possible to generate the wood grain pattern P12 that also expresses the depth of each conduit groove G.

§5. Cutting using wrinkled cutting surface In FIG. 9, a three-dimensional tree model M is cut along a plane J
An example is shown in which a wood grain pattern P12 is obtained by cutting with. However, in practice, it is preferable to use a curved surface having a natural fluctuation as the cut surface J instead of using a flat surface. In this embodiment, a curved surface (referred to as a wrinkled surface here) generated by using the scalar field of the fractal lattice is used as the cutting surface.

It is generally known that a fractal figure is a figure suitable for expressing many things in the natural world. The feature of this fractal figure is that its complexity is always the same whether it is microscopic or macroscopic. The shapes of coastlines, the shapes of trees and veins, the shapes of snowflakes, etc. found in the natural world are all typical of this fractal figure, and they are intricate even when viewed microscopically or macroscopically. It has a unique shape.
Such a property is generally called self-similarity. The essence of fractal theory lies in this self-similarity, and a more general extension of this theory makes it possible to consider a fractal lattice in which a given scalar value is self-similarly defined at each lattice point. For example, one-dimensional fractal grid,
The two-dimensional fractal lattice defines a predetermined scalar value for each lattice point arranged on a one-dimensional line segment or a two-dimensional plane, and gives a spatial scalar field. In this scalar field, each scalar value changes spatially with natural fluctuations. In other words, this change pattern, whether microscopically or macroscopically, always has the same complexity, that is, self-similarity. By incorporating such natural fluctuations in the grain pattern, it is possible to create a pattern having an impression closer to that of a natural tree.

Here, an example of a method for creating a two-dimensional fractal lattice will be described with reference to the drawings. This method is a method generally called “random midpoint displacement method”. Now, as shown in FIG. 12, the initial stage (hereinafter,
The grid points A, B, C, and D (indicated by double circles in the figure) are arranged at the four vertices of the square as grid points of the 0th stage), and the scalar values a, b, c, and d are respectively set. Define. Here, this square is an outline rectangle forming the outline of the finally generated two-dimensional fractal lattice.

Then, as shown in FIG.
Grid points E, F, G, and H of the first stage are defined at the midpoints of the grids, BCs, CDs, and DAs, respectively, and at the intersection of the diagonal lines of the four grid points ABCD, another first grid point is defined. Define a grid point I of a stage. Then, scalar values e, f, g, h, and i are defined for these five lattice points, respectively. This is the following arithmetic expression: e = (a + b) /2+T.RND (2) f = (B + c) /2+T.RND (3) g = (c + d) /2+T.RND (4) h = (d + a) /2+T.RND (5) i = (a + b + c + d ) /4+T.RND (6). In the following figures, the grid points in the state where the scalar value is undefined are shown by a single circle, and the grid points in the state where the scalar value is defined are shown by a double circle. Here, T is the maximum half-amplitude value of fluctuation, and RND is −
It is an arbitrary random number satisfying 1 ≦ RND ≦ + 1. in this way,
Random numbers are used to define the scalar values e to i,
Coincidental factors will influence. However, as each scalar value defined by the above-mentioned arithmetic expression, no definite value is defined. For example, as the scalar value e, the scalar values a of the adjacent grid points A and B,
It will be limited by b and the maximum half amplitude value T. That is, as shown in the above equation (2),
The value of the scalar value e is a value obtained by adding an arbitrary value (determined by a random number) within the range of −T to + T to the average value of the scalar values a and b. Therefore, the maximum half-amplitude value T is a parameter that limits the degree of fluctuation that deviates from the average value.

Thus, the grid points E, F, G, of the first stage are
After the scalar values e, f, g, h, and i for H and I can be defined, subsequently, as shown in FIG. 14, at the midpoint of each adjacent grid point and the intersection of the diagonals of the four grid points, The grid points J, K, L, M, N, ... of the second stage are defined respectively (in order to avoid complication, only grid points J, K, L, M, N are displayed with grid point names in FIG. Have been done). Then, for each of these lattice points, scalar values j, k, l, m, n, ... Are defined, respectively. This is the following arithmetic expression (only the arithmetic expressions for j to n are shown) j = (A + e) / 2 + (1/2) .T.RND (7) k = (e + i) / 2 + (1/2) .T.RND (8) l = (i + h) / 2 + (1 / 2) ・ T ・ RND (9) m = (h + a) / 2 + (1/2) ・ T ・ RND (10) n = (a + e + i + h) / 4 + (1/2) ・ T -Calculate by RND (11). It should be noted that the expressions (7) to (11) are very similar to the expressions (2) to (6), but the maximum half amplitude value T is multiplied by a coefficient of (1/2). .

Similarly, the third stage, the fourth stage, ...
If the process of the nth stage is repeatedly executed, the scalar value will be defined for a large number of grid points arranged on the two-dimensional plane.

The above processing will be explained as a general theory.
First, in the 0th stage, four grid points are defined at the four corner positions of the outer shape rectangle, and predetermined scalar values are defined at each grid point. Then, as the processing of the i-th stage, the following processing may be sequentially executed. That is, first, the minimum rectangle at the current stage that does not include the grid points defined up to the (i-1) th stage is recognized. For example, in the case of the first stage where i = 1, the rectangle AB shown in FIG.
CD is the minimum rectangle (the grid point A defined up to the 0th stage,
B, C, and D are not included inside), and in the case of the second stage of i = 2, the four rectangles AEIH and E shown in FIG.
BFI, HIGD, and IFCG are the smallest rectangles (first
It is a rectangle that does not include any of the grid points A to I defined up to the stage.

Then, grid points to be defined in the i-th stage are generated at the midpoints of the sides of the minimum rectangle and the center point of the minimum rectangle (for example, in the case of the first stage where i = 1, As shown in 13, the midpoint E of each side of the minimum rectangle ABCD,
Lattice points to be defined are generated at F, G, H and center point I). Furthermore, of these grid points, for the grid point generated at the middle point, a random number is applied to the scalar values of the two grid points defined up to the (i-1) th stage existing at the end points of the side. Gives the scalar value obtained by For example, for the grid point E shown in FIG. 13, a scalar value e obtained by applying a random number RND to the scalar values a and b of the two grid points A and B is given. In general, if the calculation method of the scalar value s1 with respect to the grid point defined as the midpoint of the adjacent grid points in the nth stage is shown by an equation, s1 = (α + β) / 2 + (1/2 ^(n-1) )・ T ・ RND (12). Here, α and β are scalar values (calculated in the (n−1) th stage) of grid points on both sides of the grid point.

On the other hand, with respect to the grid point generated at the center point of the minimum rectangle, the grid points (i
-1) Give a scalar value obtained by applying a random number to the scalar values of the four grid points defined up to the step. For example, for the grid point I shown in FIG. 13, the scalar value a, which has four grid points A, B, C, and D,
A scalar value i obtained by applying a random number RND to b, c, and d is given. Generally, if the calculation method of the scalar value s2 about the lattice point defined as the center point of the minimum rectangle in the nth stage is shown by an equation, s2 = (α + β + γ + δ) / 4 + (1/2 ^(n-1) ). It becomes T ・ RND (13). Here, α, β, γ, δ are 4 of the smallest rectangle.
It is a scalar value of a grid point at one corner (calculated in the (n-1) th stage).

The two-dimensional fractal lattice generated by such a method eventually gives a scalar field spread in a two-dimensional plane. Therefore, if the surface of this two-dimensional fractal grid is placed on a horizontal plane and each scalar value is plotted as a vertical height (elevation) on a graph, a concavo-convex pattern such as a mountain ridge structure can be expressed. It is known that such a raised structure has properties similar to the uneven pattern of the actual mountain raised structure existing in nature.
That is, the complexity of the concavo-convex structure is the same from a microscopic point of view and a macroscopic point of view, and even when a part of the concavo-convex structure is magnified with a magnifying glass, it has the same complexity. In other words, the individual scalar values distributed on the two-dimensional plane are arranged in a self-similar manner, and spatially increase or decrease with natural fluctuations.

By using such a two-dimensional fractal grid, it is possible to generate a wrinkled cut surface JJ having a natural fluctuation based on the cut surface J formed of a plane. The principle is shown below. Now, as shown in FIG. 15 (a),
Consider a cross section in which a cutting plane J to be a reference plane is cut perpendicularly to its main surface. It is, so to speak, a cut surface of the cut surface. Since the cross section J is a plane, its cross section becomes a straight line as shown in FIG. Here, the point Q on the cut surface J is
If it is moved to the position of the point QQ by moving the displacement amount d upward in the figure, at the position of this point QQ, the cutting plane J
Becomes a distorted surface. Therefore, a large number of points distributed on the cut surface J are similarly displaced vertically. However,
The displacement amount d has a random value at each point, and when the displacement amount has a negative value, the displacement amount is displaced downward in the figure. Then, the cut surface J is a wrinkled paper-like surface, that is, a wrinkled cut surface JJ. FIG. 15B is a cross-sectional view of the wrinkled cut surface JJ thus obtained.

By the way, in the above description, "the displacement amount d is set to a random value for each point and a large number of points distributed on the cut surface J are displaced", but this random value is set to two. If the scalar value defined by the two-dimensional fractal grid is used, the wrinkled cut surface JJ obtained becomes a curved surface having a height of the two-dimensional fractal field. In particular,
It suffices to associate each point on the cut surface J with each grid point of the two-dimensional fractal grid, and displace each point on the cut surface J with the scalar value defined for each grid point as the displacement amount d. The wrinkled cut surface JJ thus obtained is a wrinkled surface with natural fluctuations, such as an undulating structure in a mountainous area. Therefore, if the three-dimensional tree model M is cut by the wrinkle cut surface JJ and the pixels located on the wrinkle cut surface JJ are projected onto the cut surface J, a pixel group is formed on the projection surface. Although the grain pattern P12 is obtained, this grain pattern P12 is a pattern including fluctuations in the two-dimensional fractal field.

In the above example, when the wrinkled cutting surface JJ is generated based on the cutting surface J, a point Q on the cutting surface J is generated.
In a direction perpendicular to the cutting plane J (Fig. 15 (a),
Displacement amount d in the vertical direction in (b))
Q was defined. However, the displacement direction for obtaining the point QQ is not necessarily limited to the direction perpendicular to the cut surface J. If the displacement direction is determined in advance, it may be displaced in any direction. For example, as shown in FIG. 16, it is possible to set the displacement direction to “the direction toward the central axis C of the three-dimensional tree model M”.
In this example, when the displacement amount d is positive, the point QQ is located at a position displaced by the absolute value d in the direction perpendicular to the central axis C, as shown in the figure. When the displacement amount d is negative, the absolute value is defined in the direction away from the central axis C along this perpendicular line. The point QQ is defined at the position displaced by d. As shown in FIG. 4, in the annual ring model Mr in this embodiment, the potential value U has a distribution that periodically changes based on the distance from the central axis C. If the above is determined, the fluctuation of the potential value U due to the displacement amount d can be obtained by simple addition and subtraction. Therefore,
As an actual calculation, since the increase / decrease correction based on the displacement amount d (that is, the scalar value S obtained from the two-dimensional fractal grid) needs to be performed for each potential value U obtained by the cut surface J, the calculation load is required. Becomes lighter.

§6. Creation device 17 of the grain pattern pattern is a block diagram showing the basic configuration of the producing apparatus of the grain pattern pattern according to an embodiment of the present invention. This apparatus includes annual ring parameter setting means 10 for setting annual ring parameters necessary for generating the annual ring model Mr, conduit parameter setting means 20 for setting conduit parameters required for generating the conduit model Mf, and annual ring parameters. Based on the annual ring parameter set in the setting means 10,
The annual ring model generation means 30 for generating an annual ring model Mr composed of a set of pixels having a pixel value that periodically changes based on a distance from a predetermined central axis, and a conduit parameter set in the conduit parameter setting means 20. On the basis of this, by arranging a large number of conduits extending in a direction substantially along the central axis of the annual ring model Mr, and adding a conduit model Mf consisting of a set of pixels existing at least in the contour portion of this conduit to the annual ring model Mr. And a three-dimensional tree model generating means 40 for generating a three-dimensional tree model M.

This apparatus also includes a two-dimensional fractal grid generating means 50 for generating a two-dimensional fractal grid in which a predetermined scalar value is self-similarly defined at each grid point on a two-dimensional plane, and within a predetermined numerical range. Random number generating means 60 for generating the random number and the three-dimensional tree model M are cut by a predetermined wrinkle cutting plane JJ having fluctuations according to the scalar value of the two-dimensional fractal grid, and the wrinkle cutting plane JJ is cut. A pattern pattern extracting means 7 for extracting a pattern formed by a set of located pixels as a pattern pattern P12.
0, and further, a color map setting unit 80 for setting a color map for making the pixel value of each pixel correspond to a plurality of primary color components necessary for printing, and the grain pattern based on this color map. Color tone assigning means 90 for assigning a plurality of primary color components necessary for printing to each pixel constituting the wood grain pattern P12 extracted by the pattern pattern extracting means 70, and primary color components assigned by this color tone assigning means 90 By using the grain pattern P1
It has a printing unit 100 for performing printing of 2. As a result of printing by the printing unit 100, a printed matter 110 having a wood grain pattern can be obtained.

In this embodiment, in the annual ring parameter setting means 10, the annual ring parameters described in FIG. 2, that is, the average annual growth width D, the maximum variation Δ of the annual growth width Δ, and the past growth year year. 2 is set, and the annual ring model Mr generation means 30 is based on these annual ring parameters and the random numbers generated by the random number generation means 60.
An annual ring model Mr as shown in is generated. On the other hand, in the conduit parameter setting means 20, the conduit parameters described in FIG. 6, namely, the probability function Z (U), the actual length L of one pixel, the conduit average diameter φ, and the diameter variance σ ² are set. The three-dimensional tree model generating means 40 generates a conduit model Mf composed of a large number of conduits F based on these conduit parameters and the random numbers generated by the random number generating means 60, and adds this to the annual ring model Mr. As a result, FIG.
A three-dimensional tree model M as shown in is constructed.

Further, the two-dimensional fractal grid generating means 50
Uses a random number generated by the random number generation means 60 to generate a two-dimensional fractal grid, and uses this two-dimensional fractal grid to generate a wrinkled cutting plane JJ. The wood grain pattern pattern extracting means 70 cuts the three-dimensional tree model M with the wrinkle cut surface JJ, and thereby the wood grain pattern pattern P12.
To extract. At this time, regarding the pixels forming the conduit cross-section pattern P2, as described above, the temporary pixel value “12” is set.
The process of replacing "8" with the original pixel values 128 to 255 based on the depth is also performed. After all, of the wood grain pattern P12 output from the wood grain pattern pattern extracting means 70, the pixel values 0 to 1 are assigned to the pixels forming the annual ring pattern P1.
27 is assigned, and pixel values 128 to 255 are assigned to the pixels forming the conduit cross-section pattern P2.

The color map setting means 80 has a pixel value of 0.
A color map C (U) indicating color values made up of a combination of primary colors CMYK (or RGB) to be assigned to .about.255 is set. Color tone assigning means 90
Performs a process of assigning a color value to each pixel forming the grain pattern P12 using this color map C (U), and the grain pattern pattern to which this color value is assigned is printed by the printing unit as grain pattern image data. Given to 100. The printing unit 100 is composed of a printing plate and a device for printing, and prints on a vinyl chloride sheet, for example, based on the given woodgrain pattern image data to generate a printed matter 110 having a woodgrain pattern.

The means 10 to 90 in the apparatus shown in FIG. 17 are actually constituted by a computer and a storage device, and the above-mentioned processing processes are
It will be executed as data processing on the computer.

§7. Other Embodiments The present invention has been described above based on the illustrated embodiments, but the present invention is not limited to these embodiments and can be implemented in various modes. Some modifications will be described below.

In the above-mentioned §5, the embodiment in which the three-dimensional tree model M is cut by the wrinkled cutting plane JJ has been described. That is, in the embodiment of §5, the fractal fluctuation is included in the obtained wood grain pattern P12 by providing the fractal fluctuation on the wrinkle cut surface JJ side. On the other hand, fractal fluctuation may be provided on the side of the three-dimensional tree model M. In this case, prepare a two-dimensional fractal grid that defines a given scalar value at each grid point on the two-dimensional plane in a self-similar manner or a three-dimensional fractal grid defined at each grid point in three-dimensional space, and Distorted three-dimensional tree model MM is created by displacing each pixel forming tree model M based on the scalar value defined at each grid point of the two-dimensional fractal grid or three-dimensional fractal grid, and the distorted cubic By cutting the original tree model MM along a predetermined cutting plane J,
The grain pattern P12 may be extracted. It is also possible to cut the distorted three-dimensional tree model MM by the wrinkled cutting plane JJ.

In the above-described embodiments, the wood grain pattern P12 is printed on the medium, but the conduit cross-section pattern P2 can be further printed on the medium as an embossed concavo-convex structure. In this case, for example, replacing the pixel values 128 to 255 with depth values,
An embossing plate having an uneven structure may be created and embossing may be performed.

Alternatively, instead of performing embossing,
By using a so-called rendering method, it is also possible to form a shadow inside the conduit cross-section pattern P2 and print it on a plane. That is, considering a geometric model as shown in FIG. 11, it is possible to recognize the three-dimensional shape of the conduit groove G formed by cutting the three-dimensional tree model M, and a rendering method is applied to the inner wall surface of the conduit groove G. Can be used to map shadows. Therefore, if the value indicating the shading of the shadow is used as the pixel values 128 to 255 instead of using the depth value, the shading of the shadow is expressed inside the conduit cross-sectional pattern P2, and the plane Even with the conduit cross-section pattern printed above,
It becomes possible to express the stereoscopic effect of the conduit groove G. In order to perform rendering on the inner wall surface of the conduit groove G, predetermined light source conditions (light source position, light source brightness, light source type (point light source, line light source, surface light source), etc.) are set.
Based on the geometric model shown in 1, the direction of the normal line standing on each pixel on the inner wall of the conduit groove G is obtained in the form of a normal vector, the reflection condition on this inner wall is set, and on the inner wall of the conduit groove G. The reflected light intensity for each pixel may be calculated. Then, for the pixels inside the conduit cross-section pattern P2, if the pixel value is determined based on the reflected light intensity, the stereoscopic effect of the conduit groove G can be represented by shading.

In the above embodiment, the annual ring model Mr is used.
Although the coaxial cylindrical model was used as the above, it is also possible to use the coaxial conical model so that the thickness gradually decreases as it goes from the trunk to the treetop. Similarly,
It is also possible to use a model in which a conical conduit F is arranged as the conduit model Mf. Further, in the above-mentioned embodiment, all the conduits F are arranged parallel to the central axis C,
It may be arranged to be slightly inclined from the central axis C.

[0078]

As described above, according to the method and apparatus for creating a wood grain pattern according to the present invention, a three-dimensional tree model in which an annual ring model and a conduit model are fused is constructed, and the three-dimensional tree model is constructed by a predetermined cut surface. Since the wood grain pattern pattern is created by cutting, it is possible to artificially generate a natural wood grain pattern pattern close to that of natural wood, or a wood grain pattern pattern with high design that does not exist in natural wood. .

[Brief description of drawings]

FIG. 1 is a diagram illustrating a general concept of forming a grain pattern P12 by superimposing an annual ring pattern P1 and a conduit cross-section pattern P2.

FIG. 2 is a conceptual diagram showing an example of an annual ring model Mr used in the method for creating a grain pattern according to the present invention.

FIG. 3 is a partial cross-sectional view of the annual ring model Mr shown in FIG.

FIG. 4 is a diagram illustrating a periodic potential defined for the annual ring model Mr shown in FIG.

5 is a graph showing an example of a color map C (U) that associates a gradation value C with a potential value U defined for the annual ring model Mr shown in FIG.

FIG. 6 is a conceptual diagram showing an example of a three-dimensional tree model M used in the method for creating a grain pattern according to the present invention.

7 is a partial cross-sectional view of the three-dimensional tree model M shown in FIG.

FIG. 8 is a graph showing an example of a correspondence relationship between a growth time t and a conduit density D for a general tree.

FIG. 9 is a diagram illustrating the principle of generating a wood grain pattern by cutting the three-dimensional tree model M along the cutting plane J in the present invention.

10 is a diagram showing an example of a wood grain pattern P12 obtained by the principle shown in FIG.

FIG. 11 is a diagram illustrating the principle of calculating the depth for each point inside the conduit cross-section pattern P2 shown in FIG.

FIG. 12 is a diagram showing a state of the 0th stage for generating a two-dimensional fractal lattice.

FIG. 13 is a diagram showing a state of the first stage in which a two-dimensional fractal lattice is generated.

FIG. 14 is a diagram showing a second stage state in which a two-dimensional fractal lattice is generated.

FIG. 15 is a cross-sectional view showing the principle of forming a wrinkled cut surface JJ based on the cut surface J.

FIG. 16 is another cross-sectional view showing the principle of forming a wrinkled cut surface JJ based on the cut surface J.

FIG. 17 is a block diagram showing a basic configuration of a grain pattern creating apparatus according to an embodiment of the present invention.

[Explanation of symbols]

 10 ... Annual ring parameter setting means 20 ... Conduit parameter setting means 30 ... Annual ring model generating means 40 ... Three-dimensional tree model generating means 50 ... Two-dimensional fractal grid generating means 60 ... Random number generating means 70 ... Wood pattern pattern extracting means 80 ... Color map Setting unit 90 ... Color tone allocating unit 100 ... Printing unit 110 ... Printed matter of wood grain pattern C ... Center axis, gradation value C (U) ... Color map D1 to D4 ... Growth width for one year d ... Displacement amount F ... Conduit f … Maximum amount of displacement G… Conduit groove J, J1, J2… Cutting surface JJ… Wrinkled cutting surface K1 to K4… Pixel M… Three-dimensional tree model Mf… Conduit model Mr… Annual ring model P… Single conduit section Pattern P1 ... Annual ring pattern P2 ... Conduit cross-section pattern P12 ... Wood grain pattern Q, QQ ... Point on surface RND ... Random number T1 ... Natural tree T2 ... Natural tree Conduit t ... Growth period U ... Potential value Z ... Probability value Z0-Z4 ... Depth Z (U) ... Probability function

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Yasuhiro Hayashi 1-1-1, Ichigaya-Kagacho, Shinjuku-ku, Tokyo Within Dai Nippon Printing Co., Ltd. (72) Ieharu Hashizume, 1-chome, Ichigaya-Kagacho, Shinjuku-ku, Tokyo No. 1-1 Dai Nippon Printing Co., Ltd. (72) Inventor Toshio Ariyoshi 1-1-1 Ichigaya Kaga-cho, Shinjuku-ku, Tokyo Inside Dai-ni Printing Co., Ltd. (72) Inventor Yu Okamoto Ikaya-kaga, Shinjuku-ku, Tokyo 1-1-1 Dai Nippon Printing Co., Ltd. (72) Inventor Yoshio Sukegawa 311 Takemazawa, Miyoshi-cho, Iruma-gun, Saitama Dai-Nihon Total Process Building Materials Co., Ltd.

Claims (9)

[Claims] 1. A method for artificially creating a wood grain pattern having a conduit cross section, comprising a growth ring model consisting of a set of pixels having pixel values that periodically change based on a distance from a predetermined central axis. In the step of creating, and in this annual ring model, a conduit tube extending in a direction substantially along the central axis is arranged, and a conduit model consisting of a set of pixels existing at least in the contour part of the conduit tube is added to the annual wheel model. And a step of creating a three-dimensional tree model by doing, and when this three-dimensional tree model is cut by a predetermined cutting plane, a pattern composed of a set of pixels located on the cutting plane is extracted as a woodgrain pattern. A method for producing a wood grain pattern having a conduit cross section, which comprises: 2. The method according to claim 1, wherein, for each pixel existing in a region outside the conduit, a pixel value within a predetermined range is defined, and for each pixel existing in a region inside the conduit, Creation of a woodgrain pattern having a conduit cross-section characterized by defining a temporary pixel value outside the range and performing a process of replacing the temporary pixel value with a pixel value based on the depth from the cut surface to the contour of the conduit. Method. 3. The method according to claim 1, wherein the three-dimensional shape of the conduit groove formed by cutting the three-dimensional tree model is recognized, and a predetermined light source condition is set, so that each part of the conduit groove is removed. A method of creating a woodgrain pattern having a conduit cross section, wherein the reflected light intensity is calculated and the pixel value of each pixel existing in the region inside the conduit is determined by the reflected light intensity. 4. The method according to claim 1, wherein the arrangement density of the conduits is determined on the basis of the distribution of pixel values of the annual ring model. How to make. 5. The method according to claim 1, wherein a two-dimensional fractal grid in which a predetermined scalar value is self-similarly defined at each grid point on a two-dimensional plane is prepared. A wrinkle cut surface is generated by displacing each point on the reference plane in a predetermined direction according to the scalar value of each grid point of the fractal grid, and the three-dimensional tree model is cut by this wrinkle cut surface. A method of creating a woodgrain pattern having a conduit cross-section, characterized by extracting a woodgrain pattern. 6. The method according to claim 1, wherein a predetermined scalar value is defined in a self-similar manner at each lattice point on a two-dimensional plane, or in a two-dimensional fractal lattice or in a three-dimensional space. Prepare a 3D fractal grid defined at grid points, and displace each pixel that constitutes a 3D tree model based on the scalar value defined at each grid point of the 2D fractal grid or the 3D fractal grid. By creating a distorted three-dimensional tree model, and by cutting the distorted three-dimensional tree model by a predetermined cutting plane, a wood grain pattern pattern is extracted, and a method of creating a wood grain pattern pattern having a conduit cross section. . 7. An apparatus for artificially creating a woodgrain pattern having a conduit cross section, which comprises a growth ring model consisting of a set of pixels having pixel values that periodically change based on a distance from a predetermined central axis. An annual ring model generating means for generating and a conduit model, which is formed by arranging a conduit extending in a direction substantially along the central axis in the annual ring model, and including a set of pixels existing at least in a contour portion of the conduit, is defined as the annual ring model. A three-dimensional tree model generation means for generating a three-dimensional tree model by adding it to the model, and a two-dimensional fractal grid that self-reciprocally defines two-dimensional fractal grids defined at each grid point on the two-dimensional plane. The fractal grid generating means and the three-dimensional tree model are cut by a predetermined wrinkled cutting surface having a fluctuation corresponding to a scalar value of the two-dimensional fractal grid. An apparatus for creating a wood grain pattern having a conduit cross section, characterized by comprising: a wood grain pattern pattern extracting means for extracting a pattern formed by a set of pixels located on the wrinkle cut surface as a wood grain pattern pattern. . 8. The generating device according to claim 7, further comprising a random number generating unit that generates a random number within a predetermined numerical range, and the two-dimensional fractal grid generating unit uses the random number generated by the random number generating unit. Then, a two-dimensional fractal grid is generated, the annual ring model generation means defines the length of each cycle using the random numbers generated by the random number generation means, and the three-dimensional tree model generation means An apparatus for creating a wood grain pattern having a conduit cross section, characterized in that the conduits are arranged by using the generated random numbers. 9. The creating apparatus according to claim 7, further comprising: a color map setting unit that sets a color map for making a pixel value of each pixel correspond to a plurality of primary color components necessary for printing, Based on the color map, for each pixel forming the woodgrain pattern extracted by the woodgrain pattern extraction means,
A color tone assigning unit that assigns a plurality of primary color components required for printing, and a printing unit that prints the wood grain pattern using the primary color components assigned by the color tone assigning unit. An apparatus for creating a wood grain pattern having a conduit cross section.