JP4736239B2 - Pattern image creating method and apparatus - Google Patents

Description

[0001]
[Industrial application fields]
The present invention relates to a technique for creating an image by projecting an object represented in a three-dimensional space onto a two-dimensional plane, and in particular, a pattern for artificially creating a pattern such as artificial marble using such a technique. The present invention relates to an image creation method and apparatus.
[0002]
[Prior art]
Conventionally, various patterns are used to decorate the surface of building material products. Among them, the artificial marble pattern has a high-class feeling and is very popular. In order to create such a pattern of artificial marble, it has been carried out through work such as direct photographing of real artificial marble, but there is a problem that production costs are high and there are no variations . Recently, attempts have been made to artificially generate a marble pattern by computer graphics, but it is difficult to obtain an actual one.
[0003]
In general, the pattern of artificial marble is such that other objects such as pebbles are included in the medium. Since only this surface is visible from the outside, it is thought that the artificial marble pattern can be approximated by an image obtained by projecting an object like a pebble on the surface. Therefore, a technique for projecting an object in a three-dimensional space, that is, a three-dimensional CG (computer graphics) expression technique is used.
[0004]
Here, a three-dimensional CG expression technique will be described. In the three-dimensional CG representation technique, a projection method that takes into account the depth-of-field effect of an object that exists in a three-dimensional space is used in order to express more stereoscopic effect. Depth of field is the distance from the plane in which the viewpoint is in focus, and the depth of field effect refers to the depth of field in an object other than the depth of field in three-dimensional space. This is a technique that gives a sense of realism by blurring according to the difference between the two.
[0005]
A technique for rendering using such a depth-of-field effect is generally realized using a perspective radiation shadow. Here, rendering using a perspective radiation shadow will be described. FIG. 11 is a diagram illustrating the relationship between an object (object), a focal plane, and a viewpoint when performing perspective radiation. As shown in FIG. 11A, a predetermined angle of view is set for the viewpoint, and an object included in the range of the angle of view is projected onto the viewpoint. In FIG. 11A, the focal plane and the viewpoint seem to form a triangle, but since it is actually a three-dimensional space, a cone or pyramid shape with the focal plane as the bottom and the viewpoint as the apex is formed. It is a constituent relationship. The perspective radiation shadow is a method of reflecting color information at each point on a plane parallel to the focal plane in a circular or polygonal shape. By applying this viewpoint to each pixel position, a projection image is obtained.
[0006]
As described above, a projection image is obtained by performing a perspective radiation shadow on each pixel. Actually, in order to improve the accuracy, as shown in FIGS. 11B and 11C, a projection image in a state where the viewpoint position is changed (jitter processing) is obtained, and an average value of these projection images is obtained. Thus, a final projection image is obtained. As the viewpoint position is increased, a more accurate projection image is obtained. By using such a three-dimensional CG expression technique, it is possible to easily create an artificial marble pattern.
[0007]
[Problems to be solved by the invention]
As described above, when creating a marble pattern image by rendering an image of a predetermined size, the technique of the perspective radiation shadow subjected to the jitter processing has a great effect. However, when creating a pattern image or the like, it is common to create a reference unit image and obtain a pattern image of a desired size by repeatedly arranging the unit image vertically and horizontally. . In this way, when the unit image is repeated to create an image of a target size, it is preferable to perform endless processing so that the joint is not noticeable.
[0008]
However, when an endless image in which the same pattern appears repeatedly is generated, in the perspective radiation method, the visual volume (the volume in a three-dimensional space that affects one point on the generated two-dimensional image) is as shown in FIG. Therefore, even if the object arranged in the three-dimensional space is endless, the edges (up and down, left and right) of the rendered image are not connected. For this reason, a preferable endless image cannot be obtained. Also, there is no individual pattern on the surface of the projected object, and the resulting artificial marble pattern is boring.
[0009]
In view of the above points, the present invention can create a pattern image subjected to endless processing by expressing the depth-of-field effect using an orthogonal projection, and has a unique design. It is an object of the present invention to provide a pattern image creating method and apparatus capable of creating a pattern image.
[0010]
[Means for Solving the Problems]
  In order to solve the above problems, in the present invention, a computer generates a fragment in a three-dimensional space based on an input parameter, and pastes the input image on the generated fragment, and then the image is pasted. A projected image is created by projecting the debris onto a two-dimensional plane using the orthogonal projection method and calculating the projected pixel value considering the distance from the focal plane and the transparency of the mediumThe projection image is created by performing rendering a plurality of times by changing the line-of-sight direction and calculating the average of the rendering results obtained by each rendering.It is characterized by doing so. In the present invention, in particular, a projection image is obtained by projecting a fragment generated in a three-dimensional space onto a two-dimensional plane by an orthogonal projection method in consideration of the distance from the focal plane and the transmittance of the medium. Therefore, by repeating this projection image as a unit image, it becomes possible to create a pattern image with inconspicuous joints, and an artificial marble-like pattern image with a unique design pattern on the surface of the fragment It becomes possible to create.
[0011]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a flowchart of a pattern image creating method according to the present invention. First, parameters relating to how fragments are generated in the three-dimensional space are input (step S1). Such parameters include the projection area including the screen size and position, the number of texture groups, the surface texture, the size of the fragments (objects), the shape of the bottom surface of the fragments, the height of the vertices, the arrangement density, and the fragments in the projection area. Set the arrangement range of. The projection area is a three-dimensional area where the unit image is projected by orthographic projection. Here, the size of the screen S that becomes a unit image after projection is assumed to be w pixels (x direction) × h pixels (y direction). Further, the z direction indicates the depth. An example of such a projection area is shown in FIG. The number of texture groups indicates the number of types of texture on the surface of the fragment. For each texture group, the surface texture, the size of the fragment, the shape of the bottom surface of the fragment, the height of the vertex, the arrangement density, and the projection area. The arrangement range of the fragments at is set.
[0012]
The surface texture is the texture of the generated fragment surface, and is set by designating a texture for mapping to the fragment surface. The designation of the texture is performed by designating the ID of a file storing the texture as an image.
[0013]
The three parameters of fragment size, fragment bottom shape, and apex height are for determining the fragment shape. Here, an example of a fragment shape is shown in FIG. The example of FIG. 3 shows a case where the parameter of the shape of the bottom surface of the shard is set to be square. The size of the fragment is a parameter for setting the size of the bottom surface, and the height of the vertex is a parameter for setting the height of the vertex from the bottom surface in the quadrangular pyramid. In the case of the fragment shape as shown in FIG. 3, since the bottom surface is a quadrangle, it can be composed of at least two triangular polygons, and since the side surface is triangular, it can be composed of at least one triangular polygon. That is, the shape of the quadrangular pyramid shown in FIG. 3 can be composed of at least five triangular polygons.
[0014]
The arrangement range is a parameter for setting a range in which fragments are arranged in the projection area shown in FIG. This arrangement range is set by the z coordinate. FIG. 2B shows an arrangement range when the z coordinate is set to a predetermined range. In FIG. 2B, the inside of the shaded rectangular parallelepiped is the arrangement range. The arrangement density is a parameter that sets the density of the pieces arranged within the arrangement range.
[0015]
Subsequently, parameters relating to how to perform the rendering process for projecting the fragments in the three-dimensional space onto the two-dimensional plane are input (step S2). Such parameters include focal plane placement position, line-of-sight direction setting number N, line-of-sight movement maximum angle θ._maxSet. Further, since the focal plane is arranged in parallel with the screen S, the arrangement position of the focal plane is set by the distance from the screen S. The line-of-sight direction setting number N is the number of line-of-sight direction settings for performing projection, and projection is performed from different directions by the set number to determine the pixel value on the screen S. Maximum line-of-sight movement angle θ_maxIndicates the maximum angle at which the line-of-sight direction can be tilted when the line-of-sight direction and the screen S plane are perpendicular to each other.
[0016]
Next, fragments are generated in the three-dimensional space (step S3). This is generated for each texture group with the density set in the arrangement area where the pieces of the set shape are set. Also, the specified texture image is mapped, that is, pasted to each polygon constituting the generated fragment.
[0017]
A texture image mapping method will be described with reference to FIG.
First, as the texture image, an image larger than the polygon constituting the generated fragment is prepared. In the example of FIG. 4, the texture image has a size of vertical IY × horizontal IX. First, for a certain polygon, one point A (x, y) is randomly selected as shown in FIG. This is done by generating random numbers in the range of 0 ≦ x ≦ IX and 0 ≦ y ≦ IY. With respect to the selected point A (x, y), the positional relationship between the points B and C on the texture image is determined as shown in FIG. When both the points B and C fit on the texture image, the pattern determined by the three points ABC is cut out and pasted on the polygon. As shown in FIG. 4C, when either point B or point C protrudes from the texture image area, positioning of point A is performed again. The pattern to be pasted is determined for all the polygons generated as described above.
[0018]
Next, a medium polygon is generated (step S4). The medium polygon has the same size as the screen S and is generated at the same position as the screen S. That is, since the screen S has a rectangular shape, the screen S can be composed of two triangular polygons. A medium texture prepared in advance can be attached to the medium polygon. However, in this embodiment, in order to make the medium uniform, the medium polygon is set to have a uniform color over the entire surface.
[0019]
Next, orthographic projection of the fragments on the screen S is performed to determine the color of each pixel on the screen (step S5). A plurality of line-of-sight directions for orthographic projection in step S5 are set, and a pixel value on the screen S is determined for each line-of-sight direction. The pixel values on the screen S obtained for each line-of-sight direction are finally averaged to obtain pixel values constituting one projection image. Here, a case where the line-of-sight direction from the screen S is perpendicular to the screen S will be described first.
[0020]
In this case, a pixel on a certain fragment is projected onto a pixel on the screen S having the same (x, y) coordinate value. At this time, the pixel value V of the pixel on the screen is displayed._HIs calculated by the following (Formula 1).
[0021]
(Formula 1)
V_H = T x (D_O / D_F) × V_O
T: Decay rate
[0022]
However, in the above (Formula 1), D_OIs the pixel V on the shard from the screen S_ODistance to D_FIndicates the distance from the screen S to the focal plane. The above (Equation 1) indicates that the closer the pixel on the fragment is to the focal plane, the closer to the original pixel value, and the farther away from the focal plane, the more blurred the pixel. Further, since the fragments are randomly arranged in the three-dimensional space, only the fragments closest to the screen S can be seen with respect to the fragments overlapping on the line of sight from the screen S. This can be expressed by simply comparing the z-coordinates of both pieces and utilizing the one close to the screen S.
[0023]
Furthermore, the pixel value on the screen S is not determined only by this, but is determined in consideration of the attenuation due to the medium. Set the pixel value of the medium to V_BAssuming that the transmittance for synthesis is α (where 0 ≦ α ≦ 1), the pixel value V on the screen S is calculated by the following (Formula 2). At this time, the pixel value V of the medium_BThe pixel value on the medium polygon generated on the screen is adopted.
[0024]
(Formula 2)
V = α x V_H + (1-α) x V_B
[0025]
As can be seen from (Equation 2), the larger the value of the transmittance α, the more the original color is expressed, and the smaller the value of the transmittance α, the stronger the influence of the medium color.
[0026]
The projection is performed in the same manner when the viewing direction is changed. For example, when the orthogonal projection is performed at the line-of-sight angle θ, the line-of-sight angle θ is calculated by the following (Formula 3).
[0027]
(Formula 3)
θ = θ_max  × γ₁
[0028]
In the above (Formula 3), θ_maxIndicates the maximum line-of-sight movement angle set in step S1, and γ₁Indicates a random number generated to have a value in the range of 0.0 to 1.0.
[0029]
Based on this viewing angle θ, the pixels on the screen S with respect to the viewing angle θ can be obtained by orthogonally projecting the fragments present in the three-dimensional space and considering the influence of the medium. Here, FIG. 5 shows the relationship between the line-of-sight angle θ, the screen S, and fragments when attention is paid to a certain fragment. In the figure, the horizontal axis is the z-axis shown in FIG.
FIG. 5A shows a state in which the viewing angle is 0 °, that is, the viewing direction is perpendicular to the screen S. FIGS. 5B and 5C show states in which the viewing direction is inclined by the viewing angle θ. Although the screens and pieces shown in FIGS. 5A to 5C are all present at the same position, since the viewing angles are different, the state of the pieces projected on the screen is different. The state of the fragments projected on the screens of FIGS. 5A to 5C are shown in FIGS. In FIG. 6A, the object is projected on the center of the screen, but in FIG. 6B, it is seen that the object is projected below the screen, and in FIG. 6C, it is projected above the screen.
[0030]
Orthographic projection in step S5 is performed for the set number of line-of-sight directions set in step S2. That is, as many rendering results as the number of gaze direction settings are obtained. The finally obtained projection image is an average of these rendering results. Here, details of the processing procedure in step S5 will be described based on the flowchart of FIG. First, as described above, the line-of-sight angle θ is calculated using (Equation 3) (step S11).
[0031]
Subsequently, the line-of-sight vector (Vx, Vy) in the screen plane for determining the direction in which the line-of-sight angle θ is inclined is calculated by the following (Formula 4) (Step S12).
[0032]
(Formula 4)
Vx = sinφ
Vy = cosφ
However, φ = 360 ° × γ₂
[0033]
In the above (Equation 4), γ₂Indicates a random number generated to have a value in the range of 0.0 to 1.0. In this way, when the line-of-sight angle θ and the line-of-sight vector (Vx, Vy) are determined, the line-of-sight direction for performing the orthogonal projection is specified as one direction, and thus the orthogonal projection from the specified line-of-sight direction to the screen, That is, rendering on the screen is performed. Specifically, this is performed using the above (Formula 1) and (Formula 2), but when the line-of-sight direction θ is not 0 °, position calculation at each of the three-dimensional coordinates is necessary for calculating the distance. Therefore, the calculation load increases. Therefore, in the present invention, in order to reduce this calculation load, before the rendering process is performed, the position coordinates of the fragments are converted so that the distance can be calculated with one coordinate axis (step S13).
[0034]
Specifically, since each piece is composed of a polygon, the coordinate value of each vertex of the polygon is converted by the following (Equation 5) and (Equation 6).
[0035]
(Formula 5)
K = tan θ × abs (L−Z)
[0036]
The above (Equation 5) is for calculating the movement amount K, L indicates the z coordinate value of the projection plane, Z indicates the z coordinate value of the polygon, and abs indicates the absolute value.
[0037]
(Formula 6)
Mx = Vx × K
My = Vy × K
[0038]
The above (Equation 6) calculates the movement amounts Mx and My of the x coordinate and the y coordinate using the movement amount K calculated in (Equation 5). As a result, the coordinate values (X, Y, Z) of the vertices of each polygon are converted into virtual coordinate values (X + Mx, Y + My, Z).
[0039]
Such coordinate conversion processing in step S13 is performed for each vertex of each polygon constituting each piece. As a result, coordinate values in a virtual three-dimensional space are created. For example, when a polygon exists in a state as shown in FIG. 8A, it is converted into a state as shown in FIG. As a result, the vertex A (X, Y, Z) having the polygon constituting the fragment moves to A ′ (X + Mx, Y + My, Z). By converting in this way, the difference between the values in the z-axis direction (left-right direction in FIG. 8) becomes the distance from the projection plane as it is.
[0040]
Subsequently, the pixel value V of the pixel on the screen S is calculated by using (Equation 1) and (Equation 2) based on this virtual coordinate value. At this time, the pixel value V in (Formula 1)_HDistance D for calculation_O, Distance D_FBoth need only use z-coordinate values in a virtual space, so that the calculation load is greatly reduced and processing can be performed at high speed. This set of pixel values is the rendering result when the line-of-sight direction is set (step S14).
[0041]
The result of rendering on the screen S is recorded in the image memory (step S15). In this case, the first rendering result is written in the image memory as it is. The second and subsequent rendering results are added to a plurality of rendering results including the contents previously written in the image memory, and are written as an average value so far. For example, the second rendering result is calculated by adding the first rendering result already recorded in the image memory to each pixel and dividing by two, and calculating the average value up to the second rendering average. Update the contents of the image memory. The result of the third rendering is the result of adding the average of rendering up to the second to twice and each pixel, then dividing by 3 to calculate the average up to the third time, and then rendering up to the third Update the contents of the image memory as an average. When this is generalized, the nth rendering result is obtained by adding the average of rendering up to the (n-1) th to (n-1) times for each pixel and then dividing by n until the past nth. And the content of the image memory is updated by using the average value of n as the rendering average up to the nth time. That is, whenever the processing in step S15 ends, the average of the rendering results so far is recorded in the image memory.
[0042]
When the average value of the rendering results is recorded in step S15, it is determined whether or not the number of times of performing the processing from step S11 to step S15 has reached the line-of-sight angle setting number N (step S16). If it has been executed N times, the process ends. If it has not been executed N times, the process returns to step S11 to continue the process. If YES is determined in the step S15, the rendering result so far recorded in the image memory is used as the projection image.
[0043]
Through the processing described with reference to the flowchart of FIG. 7, the projection image creation processing by orthographic projection in step S5 is executed. Thereby, a desired projection image is obtained.
[0044]
In the present invention, even when the size of the image memory for rendering is limited, the invention is devised so that highly accurate rendering is possible. Next, such a case will be described with reference to the flowchart of FIG. First, the rendering range is divided (step S21). This depends on the size of the unit image to be created, but here, it is assumed that 4 divisions of 2 × 2 are performed. As described in the example of FIG. 2, when the size of the screen S is w pixels (x direction) × h pixels (y direction), one rendering range is w / 2 pixels (x direction) × h / 2 pixels ( y direction).
[0045]
The subsequent processing from step S22 to step S24 is performed for each of the divided rendering ranges. First, marking is performed on a polygon of a fragment completely included in one rendering range (step S22). Further, another marking is performed on the fragments included in the rendering range and included in the other rendering ranges, that is, the polygons of the fragments that fall on the boundaries of the plurality of rendering ranges (step S23). The determination as to whether the image is included in the rendering range in step S22 or step S23 or whether it falls on the boundary is the maximum line-of-sight movement angle θ_maxIs taken into consideration. That is, the maximum line-of-sight movement angle θ_maxEven if the line of sight is moved, it is marked in step S22 in the rendering range, and the line-of-sight movement maximum angle θ is marked._maxIf there is a case where the line of sight is moved and there is a case where it is out of the rendering range, marking is performed in step S23.
[0046]
Subsequently, a divided projection image creation process is performed (step S24). As this specific process, the same process as described in the flowchart of FIG. 7 is performed. When one divided projection image is obtained, this image is stored in another storage area to make the image memory usable, and the process returns to step S22 to perform the same processing for the other rendering ranges. However, the processing in step S24 is different for the second and subsequent rendering ranges. In the second and subsequent cases, regarding the polygons of the fragments on the rendering boundary, those already processed are not processed again. Whether or not it has already been processed can be determined by the marking performed in step S23. As described above, when the divided projection images are obtained for the entire rendering range, the divided projection images are integrated to obtain a final projection image (step S25).
[0047]
(Device configuration)
Next, an apparatus configuration for executing the pattern image creation method will be described. FIG. 10 is a block diagram of a pattern image creating apparatus according to the present invention. In FIG. 10, the parameter input means 1 is for executing step S1 and step S2 of FIG. 1, and can be realized by a mouse, a keyboard, or the like.
[0048]
The fragment generation means 2 is for executing step S3 of FIG. 1, and has a function of arranging the fragments in the set three-dimensional space according to the parameters input from the parameter input means 1. The medium polygon generating means 3 is for executing step S4 in FIG. 1, and has a function of pasting a designated color or texture.
[0049]
The projecting means 4 is for executing step S5 in FIG. 1. While projecting the pieces subjected to the coordinate transformation processing of step S13 by the coordinate transformation means 5, the rendering result is written in the image memory 6. It has a function of repeatedly performing projection processing. Each of the fragment generation means 2, medium polygon generation means 3, projection means 4, and coordinate conversion means 5 is actually realized by a computer and a dedicated program installed in the computer, and the image memory 6 is installed inside the computer. .
[0050]
The output means 7 is for outputting a projection image obtained as a result of the processing by the projection means 4, and a display for display, FD, MO, etc. for outputting as image data can be applied.
[0051]
【The invention's effect】
As described above, according to the present invention, the necessary parameters are input, the image to be generated on the surface of the fragments is input, the fragments are generated in the three-dimensional space based on the input parameters, and the generated fragments After the image is pasted on, a projected image is created by projecting the fragment with the image pasted onto a two-dimensional plane by orthographic projection and calculating the projected pixel value considering the transparency of the medium Since this projection image is repeatedly arranged as a unit image, it becomes possible to create a pattern image with inconspicuous joints, and an artificial marble tone with a unique design pattern on the surface of the fragment There is an effect that a pattern image can be created.
[Brief description of the drawings]
FIG. 1 is a flowchart showing a pattern image creating method according to the present invention.
FIG. 2 is a diagram showing a projection area set in a three-dimensional space.
FIG. 3 is a diagram showing an example of a fragment shape.
FIG. 4 is a diagram for explaining determination of a pattern to be mapped to a polygon.
FIG. 5 is a diagram illustrating a relationship between a line-of-sight angle, a screen, and an object when performing orthographic projection.
6 is a diagram illustrating a state where the relationship illustrated in FIG. 5 is viewed from the screen side.
FIG. 7 is a flowchart showing details of the process in step S5 of FIG. 1;
FIG. 8 is a diagram for explaining the coordinate conversion processing in step S13 of FIG. 7;
FIG. 9 is a flowchart illustrating processing when the size of the image memory is limited.
FIG. 10 is a functional block diagram showing a configuration of a pattern image creating apparatus according to the present invention.
FIG. 11 is a diagram illustrating a relationship between an angle of view, a focal plane, and an object when performing perspective radiation.
FIG. 12 is a diagram illustrating a relationship between a viewpoint position, a rendering plane, and a viewing volume when performing perspective radiation.
[Explanation of symbols]
S ... Screen
1 ... Parameter input means
2 ... Fragment generating means
3 ... Medium polygon generating means
4 ... Projection means
5. Coordinate conversion means
6 ... Image memory
7 ... Output means

Claims (5)

A computer generates a fragment in a three-dimensional space based on the input parameters, a step of pasting the input image on the generated fragment, and a two-dimensional plane of the fragment on which the image is pasted A projection image creation step of creating a projection image by calculating a pixel value to be projected in consideration of the distance from the focal plane and the transparency of the medium ,
The projection image creating step is characterized in that the projection image is created by performing rendering a plurality of times by changing the line-of-sight direction and calculating an average of rendering results obtained by each rendering. Image creation method.   In the projection image creation step, the line-of-sight direction is determined based on the input parameters, and after the coordinates of the generated fragment in the three-dimensional space are converted based on the line-of-sight direction, projection is performed by an orthographic projection method. The pattern image creating method according to claim 1, wherein the pattern image creating method is a method.   In the coordinate conversion, the coordinates of the fragments in the three-dimensional space are converted into coordinates in the three-dimensional space with the distance direction from the projection plane as one coordinate axis using the line-of-sight angle and the line-of-sight vector that determine the line-of-sight direction. The pattern image creating method according to claim 2, wherein the pattern image creating method is one. Parameter input means for inputting necessary parameters and setting an image to be generated on the surface of the fragment, and generating the fragment in the three-dimensional space based on the input parameter and pasting the set image on the generated fragment A debris generation unit to be attached, a line-of-sight direction based on the input parameters, a coordinate conversion unit for converting coordinates in the three-dimensional space of the debris based on the line-of-sight direction, and the coordinate-converted debris in two dimensions while projected by orthogonal projection method in a plane, it possesses a projection means for creating a projection image by calculating a pixel value projected by considering the permeability of the distance and the medium from the focal plane, and
The projection unit includes an image memory for writing a rendering result, performs rendering a plurality of times by changing the line-of-sight direction, and averages the rendering results obtained by each rendering based on the number of times of writing to the memory so far. A pattern image creating apparatus for creating a projected image by overwriting the image memory while taking a picture . The coordinate conversion means converts the coordinates of the fragments in the three-dimensional space into coordinates in the three-dimensional space with the distance direction from the projection plane as one coordinate axis, using the line-of-sight angle and the line-of-sight vector that determine the line-of-sight direction. The pattern image creating apparatus according to claim 4 , wherein the pattern image creating apparatus is a thing.

Images