JPH0535883A - Three-dimensional image generating method - Google Patents

Abstract

PURPOSE:To obtain the three-dimensional image generating method which can eliminate processing having no direct contribution to image generation. CONSTITUTION:A processor 1 compares a depth coordinate value (z) among screen coordinates with the value in a Z buffer memory 3 in an implicit surface erasure processing step based upon Z buffer algorithm to decide the front-rear relation of a figure in picture element units and select graphic data which are closer to a point of vision, and stores the identification number of the figure in a frame buffer 2. Further, the processor 1 calculates brightness I on a screen in the picture element units in a brightness calculation step by using the identification number in the frame buffer 2 at the end of the implicit surface erasure processing step for all graphic data and stores the brightness in the frame buffer memory 2. Then the processor 1 displays a three-dimensional image on a display device 4 according to the contents of the frame buffer memory 2.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a three-dimensional image generation method for stereoscopically displaying an image using graphic data stored in a storage device of an electronic computer system.

[0002]

2. Description of the Related Art In recent years, with the widespread application of three-dimensional image generation technology such as design and simulation, a larger amount of shape data can be handled and a three-dimensional image with higher image quality can be efficiently generated. It is desired to develop a generation method. FIG. 9 is a block diagram of a three-dimensional image generation apparatus that realizes a conventional three-dimensional image generation method. This three-dimensional image generation apparatus includes a processor 31, a frame
A buffer memory 32, a Z buffer memory 33,
For example, the display device 34 is a CRT. The processor 31 reads the graphic data from a storage device (not shown) and performs arithmetic processing such as calculation of the depth coordinate value Z and the brightness I. The frame buffer memory 32 stores the brightness of pixels. The Z buffer memory 33 stores the depth coordinate value Z of the pixel. The display device 34 displays the contents stored in the frame buffer memory 32.

FIG. 10 is a flow chart for explaining the procedure of a conventional three-dimensional image generating method for performing hidden surface removal called a Z-buffer algorithm. This three-dimensional image generating method is executed by the three-dimensional image generating apparatus shown in FIG. It This Z-buffer algorithm is, for example, Nikkei Computer Graphics, December 1987 issue, p. 126, or "Basic Graphics" by Kei Kawai, Shokoido, 19
1985, p.186, etc. First, the frame buffer memory 32 and the Z buffer memory 33 are initialized (step S101). That is, the background color brightness is stored in the frame buffer memory 32 as Z
The maximum value of the depth coordinate value Z (the coordinate value of the position farthest from the viewpoint) is set in the buffer memory 33. Next, the polygons are scan-converted in an arbitrary order (step S10).
2). Next, the depth coordinate value Z is calculated for each pixel in the scan-converted polygon (step S103). Next, the calculated depth coordinate value Z is compared with the depth coordinate value ZB stored in the Z buffer memory 33 (step S104), and the calculated depth coordinate value Z is stored in the frame buffer memory 32. When it is smaller than the depth coordinate value ZB, that is, at a position closer to the viewpoint, the value of the Z buffer memory 33 is replaced with the calculated depth coordinate value Z (step S105), and the brightness of the pixel is calculated from the shading calculation (step S106). Is stored in the frame buffer memory 32 (step S107). In step S104, step S1
The depth coordinate value Z calculated in 03 is the Z buffer memory 3
When it is determined that the depth coordinate value is equal to or larger than the depth coordinate value ZB stored in 3, that is, the position is far from the viewpoint, or after the process of step S107 is performed, all pixels included in the polygon are processed in steps S103 to S1
It is determined whether or not the process of 07 is completed (step S10).
8) If not completed, return to step S103. If finished, step S10 for all polygons.
It is determined whether or not the processes of 2 to S108 are completed (step S109), and if not completed, the process returns to step S102. If finished, this routine is finished.

At the end of the above processing, only the brightness of the polygon closest to the line of sight is stored in the frame buffer memory 32 for each pixel on the screen, and the desired three-dimensional image is generated. In particular, in the calculation based on the Gouraud shading algorithm, which is one of the smooth shading algorithms for performing smooth shading, the luminance I can be sequentially calculated in the same manner as the depth coordinate value Z. It can generate a three-dimensional image at high speed.

[0005]

However, in the above-mentioned conventional method, since the shading calculation (luminance calculation) is performed before the hidden surface removal is completed, when the polygon closer to the viewpoint is processed later, the shading is performed. There is a problem that the calculation needs to be redone and the processing time that does not contribute to the image generation of the final result increases. In particular, in order to generate a three-dimensional image having higher image quality than Gouraud shading, more complicated shading calculation needs to be performed, and the amount of shape data to be handled increases. That is, the pursuit of a higher quality three-dimensional image increases the number of shading calculations that become invalid, and increases the processing time that does not contribute to image generation of the final result.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a three-dimensional image generation method capable of reducing the processing that does not directly contribute to image generation.

[0007]

According to a first aspect of the present invention, there is provided a Z buffer for selecting graphic data closer to a viewpoint by determining the front and back of a graphic for each pixel by using a depth coordinate value z of screen coordinates. After the hidden surface removal processing step based on the algorithm and the hidden surface removal processing step for all the graphic data, the brightness I is calculated for each pixel on the screen.
And a brightness calculation step for calculating.

According to a second aspect of the present invention, in the first aspect of the invention, the brightness I of the graphic is (x,
y) When it can be uniquely determined by the coordinate value, that is, can be defined as I = g (x, y) using the function g, the luminance I is calculated from the pixel position, that is, the (x, y) coordinate value in the luminance calculation step. It is characterized in that it is directly calculated using the formula I = g (x, y).

According to a third aspect of the present invention, in the first aspect of the invention, the brightness I of the graphic is (x,
y) It can be uniquely determined by the coordinate value, that is, it can be defined as I = g (x, y) using the function g, and the luminance I ′ in the pixel whose luminance value was calculated immediately before and the x coordinate or the y coordinate of the function g When I = I ′ + g ′ (x, y) can be calculated using the partial differential component g ′ (x, y) with respect to When I'is the same as the figure of the pixel for which the pixel is calculated, the brightness I is calculated using the above formula I = I '+ g' (x, y), and the figure of the pixel for which the brightness I is to be calculated has the brightness value immediately before. When it is different from the figure of the pixel, the brightness I
It is characterized in that it is calculated directly by using = g (x, y).

According to a fourth aspect of the present invention, screen coordinates (x, y, consisting of coordinates (x, y) on the two-dimensional screen plane and coordinates z representing the depth from the screen plane.
z), the attribute vector (t1, t2, ..., T) including a set of three-dimensional graphic data defined by z) and attribute data that determines the characteristics peculiar to the graphic such as color, brightness, and normal vector.
n) (n is an integer of 2 or more), a three-dimensional image generating method is used to determine the front and back of a figure for each pixel using the depth coordinate value z of the screen coordinates. Then, the hidden surface erasing processing step based on the Z-buffer algorithm for selecting the graphic data closer to the viewpoint, and after the hidden surface erasing processing step for all the graphic data is completed, the element of the attribute vector is set for each pixel on the screen. An attribute value calculation step of calculating a value of a certain attribute element ti (i is an integer from 1 to n) is performed.

According to the invention of claim 5, in the invention of claim 4, the attribute element ti is (x,
y) Can be uniquely determined by the coordinate values, that is, the function gi
When it is possible to define as ti = gi (x, y) using, the attribute value ti is calculated from the position of the pixel, that is, the (x, y) coordinate value, into the equation ti = gi in the attribute value calculation step.
The feature is that it is calculated directly using (x, y).

According to a sixth aspect of the invention, in the fourth aspect of the invention, the value of the attribute element ti can be uniquely determined by the (x, y) coordinate value of the screen coordinates, that is, using the function gi, ti = can be defined as gi (x, y),
And the value of the attribute element t'i calculated immediately before and x of the function gi
Partial differential component g'i (x, y) with respect to coordinates or y coordinates
When it is possible to calculate ti = t′i + g′i (x, y) by using, in the characteristic value calculation step, the figure of the pixel of the attribute element ti whose value is to be calculated is the attribute element ti whose value was calculated immediately before. When it is the same as the figure of the pixel of ', the value of the attribute element ti is calculated using the above expression ti = t'i + g'i (x, y), and the figure of the pixel of the attribute element ti whose value is to be calculated Is different from the calculated figure of the pixel of the attribute element ti ′, it is directly calculated using the equation ti = gi (x, y).

[0013]

According to the invention of claim 1, in the hidden surface elimination processing step based on the Z-buffer algorithm, the depth coordinate value z of the screen coordinates is used to determine the front and back of the figure for each pixel and the viewpoint is determined. Select close figure data,
In the brightness calculation step, the brightness I is calculated for each pixel on the screen after the hidden surface removal processing step for all graphic data is completed.

According to the second aspect of the invention, the luminance I of the figure is I.
Can be uniquely determined by the (x, y) coordinate value of the screen coordinates, that is, I = g (x,
If it can be defined as y), the brightness I is directly calculated in the brightness calculation step from the position of the pixel, that is, the (x, y) coordinate value using the formula I = g (x, y). In the invention of claim 3, the brightness I of the figure can be uniquely determined by the (x, y) coordinate value of the screen coordinates, that is, can be defined as I = g (x, y) using the function g, And the luminance I ′ in the pixel for which the luminance value was calculated immediately before
And the partial differential component g'of the function g with respect to the x coordinate or the y coordinate
When I = I ′ + g ′ (x, y) can be calculated using (x, y), the figure of the pixel for which the brightness I should be calculated is the pixel whose brightness I ′ was calculated immediately before in the brightness calculation step. When it is the same as the figure, the brightness I is expressed by the formula I = I '+ g' (x,
y), and when the figure of the pixel for which the brightness I is to be calculated is different from the figure of the pixel for which the brightness value was calculated immediately before, the brightness I is directly calculated using the formula I = g (x, y). To do.

According to the invention of claim 4, the Z buffer
In the hidden surface removal processing step based on the algorithm, the depth coordinate value z of the screen coordinates is used to determine the front and back of the figure for each pixel and select figure data closer to the viewpoint. After the hidden surface erasing processing step for the figure data of (3) is finished, the value of the attribute element ti (i is an integer from 1 to n) which is an element of the attribute vector is calculated for each pixel on the screen.

In the invention of claim 5, the attribute element ti
Can be uniquely determined by the (x, y) coordinate value of the screen coordinates, that is, ti = gi using the function gi.
If it can be defined as (x, y), in the attribute value calculation step, the attribute element ti is directly calculated from the position of the pixel, that is, the (x, y) coordinate value using the formula ti = gi (x, y). To do.

In the invention of claim 6, the attribute element ti
Can be uniquely determined by the (x, y) coordinate value of the screen coordinates, that is, ti =
can be defined as gi (x, y) and can be defined as ti = t using the value of the attribute element t'i calculated immediately before and the partial differential component g'i (x, y) with respect to the x coordinate or the y coordinate of the function gi. '
When it is possible to calculate i + g′i (x, y), when the figure of the pixel of the attribute element ti whose value is to be calculated is the same as the figure of the pixel of the attribute element ti ′ whose value was calculated immediately before in the characteristic value calculation step. The value of the attribute element ti is expressed by the expression ti = t'i +
g'i (x, y) is used to calculate the value, and the attribute element t for which the value of the pixel of the attribute element ti is calculated immediately before
When it is different from the figure of the pixel of i ′, the expression ti = gi (x,
Calculate directly using y).

[0018]

Embodiments of the present invention will now be described in detail with reference to the drawings. (Embodiment 1) FIG. 1 is a block diagram of a three-dimensional image generating apparatus which realizes a three-dimensional image generating method according to a first embodiment of the present invention. This three-dimensional image generating apparatus includes a processor 1, a frame buffer memory. 2 and Z buffer memory 3
And a display device 4 including, for example, a CRT. The processor 1 reads graphic data from a storage device (not shown) and performs arithmetic processing such as calculation of the depth coordinate value Z and the brightness I. The frame buffer memory 2 stores the luminance of pixels. The Z buffer memory 3 stores the depth coordinate value Z of the pixel. The display device 4 is a frame
The contents stored in the buffer memory 2 are displayed.

FIG. 2 is a flow chart for explaining the procedure of the three-dimensional image generating method according to the first embodiment of the present invention. This three-dimensional image generating method is executed by the three-dimensional image generating apparatus shown in FIG. A basic figure such as a triangle or a quadrangle is called a polygon, and each polygon is given a positive positive number of 1 or more as an identification number. First, the frame buffer memory 2 and the Z buffer memory 3 are initialized (step S1). That is, 0 is set as the polygon identification number (hereinafter referred to as "polygon id") in the frame buffer memory 2, and the maximum value of the depth coordinate value Z (the coordinate at the position farthest from the viewpoint is set in the Z buffer memory 3). Value). Next, the polygons are scan-converted in the order of polygon id (step S2). Next, the depth coordinate value Z is calculated for each pixel in the polygon (step S3). Next, the calculated depth coordinate value Z and Z buffer
The depth coordinate value ZB stored in the memory 3 is compared (step S4), and when the calculated depth coordinate value Z is smaller than the depth coordinate value ZB stored in the Z buffer memory 3, that is, closer to the viewpoint. If in position,
The value in the Z buffer memory 3 is replaced with the calculated depth coordinate value Z (step S5), and the polygon id of the polygon is stored in the frame buffer memory 2 (step S6). When it is determined in step S4 that the depth coordinate value Z calculated in step S3 is equal to or larger than the depth coordinate value ZB stored in the Z buffer memory 3, that is, when the position is far from the viewpoint, or the process of step S6. After completion of the above step, it is judged whether or not the steps S3 to S6 have been completed for all the pixels included in the polygon (step S7), and if not completed, the procedure returns to step S3. If it is completed, it is determined whether or not the hidden surface removal processing of steps S2 to S7 is completed for all polygons (step S8). If not completed, the processing returns to step S2. If completed, shading calculation (luminance calculation) is performed for each pixel and stored in the frame buffer memory 2 (step S9), and this routine is completed.

FIG. 3 is a flow chart for explaining the procedure of shading calculation (luminance calculation), which corresponds to step S9 of FIG. The coordinates (x,
In y), both x and y are represented by positive integer values only. First, the y coordinate of the coordinates (x, y) on the screen indicating the position of the pixel to be processed is initialized to 1 (step S11). Next, the coordinates (x,
The x coordinate of y) is initialized to 1 (step S1)
2). Next, the polygon id at the coordinate (x, y) on the screen is read from the frame buffer memory 2 (step S13). Next, the function g for brightness calculation designated by the polygon id is selected (step S14). Next, the brightness I is calculated by the formula I = g (x, y) using the function g and the coordinates (x, y) on the screen (step S1).
5). That is, shading calculation (luminance calculation) is performed. Next, the calculated brightness I is stored in the position (x, y) of the frame buffer memory 2 (step S16).
Next, the x coordinate is incremented by 1 (step S17). Next, the new x coordinate value is compared with the maximum x coordinate value Xmax on the screen (step S18). If the new x coordinate is less than or equal to Xmax, the process returns to step S13. New x
If the coordinate is larger than Xmax, the y coordinate is incremented by 1 (step S19). Next, the new y-coordinate value is compared with the maximum y-coordinate value Ymax on the screen (step S20). If the new y-coordinate is less than Ymax, the process returns to step S12. If the new y coordinate is larger than Ymax, this routine ends.

As described above, since the shading calculation is performed only on the pixels corresponding to the final image after the hidden surface removal processing, the invalid shading calculation can be eliminated. For example, for g (x, y), 1 for (x, y)
By using the following equation, it is possible to obtain the brightness based on the Gouraud shading algorithm, and it is possible to obtain an image of the same image quality as the conventional method in a short time. It is obvious that the effect of reducing the number of processes according to the present invention becomes more remarkable as the number of polygon data to be handled increases. Second Embodiment In the first embodiment, the shading calculation (luminance calculation) is performed according to the procedure of FIG. 3, but the shading calculation (luminance calculation) may be performed according to the procedure of FIG. That is, first, the y coordinate of the coordinates (x, y) on the screen indicating the position of the pixel to be processed is initialized to 1 (step S21). Next, similarly, the x coordinate of the coordinates (x, y) on the screen is initialized to 1 (step S22). Next, the polygon id of the coordinate (x, y) on the screen is read from the frame buffer memory 2 (step S23). Next, the read polygon id is compared with the polygon id of the adjacent pixel (step S2
4), the read polygon id and the polygon i of the adjacent pixel
A function g (x, y) for calculating the brightness I when d is equal to
Value g'at the pixel (x, y) of the partial differential component in the x direction of
(X, y) is calculated (step S25). Next, the luminance I ′ at the pixel (x−1, y) adjacent to the pixel (x, y)
And g ′ (x, y), the luminance I of the pixel (x, y)
Is calculated by the formula I = I ′ + g ′ (x, y) (step S
26). When it is determined in step S24 that the polygon id read in step S23 is not equal to the polygon id of the adjacent pixel, the function g for brightness calculation designated by the read polygon id is selected (step S27). Next, the function g and the coordinates (x, y) on the screen
The luminance I is calculated by the equation I = g (x, y) using and (step S28). Next, the brightness I calculated in step S26 or step S28 is stored in the position (x, y) of the frame buffer memory 2 (step S29).
Next, the x coordinate is incremented by 1 (step S30). Next, the new x coordinate value is compared with the maximum x coordinate value Xmax on the screen (step S31). If the new x coordinate is less than or equal to Xmax, the process returns to step S23. New x
If the coordinate is larger than Xmax, the y coordinate is incremented by 1 (step S32). Next, the new y-coordinate value is compared with the maximum y-coordinate value Ymax on the screen (step S33). If the new y-coordinate is less than Ymax, the process returns to step S22. If the new y coordinate is larger than Ymax, this routine ends.

As described above, if the partial differential component g '(x, y) of the function g (x, y) in the x direction is used, the calculation becomes easy and the processing speed can be increased. The partial differential component of the function g (x, y) in the y direction may be used. (Third Embodiment) FIG. 5 is a block diagram of a three-dimensional image generating apparatus which realizes a three-dimensional image generating method according to a third embodiment of the present invention. This three-dimensional image generating apparatus includes a processor 11 and a frame buffer memory. 12, a Z buffer memory 13, a display device 14 such as a CRT, and an attribute buffer memory 15. The processor 11 reads graphic data from a storage device (not shown),
Calculation processing such as calculation of the depth coordinate value Z and the attribute element ti is performed. The frame buffer memory 12 stores the brightness of pixels. The Z buffer memory 13 stores the depth coordinate value Z of the pixel. The display device 14 displays the contents stored in the frame buffer memory 12. The attribute buffer memory 15 stores the attribute vector. FIG. 6 is a flow chart for explaining the procedure of the three-dimensional image generation method in the third embodiment of the present invention. This three-dimensional image generation method is the three-dimensional image generation apparatus of FIG. Executed by. As in the first embodiment, it is assumed that each polygon is given a positive positive number of 1 or more as an identification number. First the frame
The buffer memory 12 and the Z buffer memory 13 are initialized (step S41). That is, the frame
In the buffer memory 12, 0 is set as the polygon id and Z
The maximum value of the depth coordinate value Z (the coordinate value of the position farthest from the viewpoint) is set in the buffer memory 13. Next, the polygons are scan-converted in the order of the polygon id (step S42). Next, the depth coordinate value Z is calculated for each pixel in the polygon (step S43). Next, the calculated depth coordinate value Z is compared with the depth coordinate value ZB stored in the Z buffer memory 13 (step S44), and the calculated depth coordinate value Z is stored in the Z buffer memory 13. If the coordinate value is ZB or more, that is, if the position is far from the viewpoint, the process proceeds to step S47.
When the calculated depth coordinate value Z is smaller than the depth coordinate value ZB stored in the Z buffer memory 13, that is, when the position is closer to the viewpoint, the value of the Z buffer memory 13 is set to the calculated depth coordinate value Z. Replace (step S45). Next, the frame buffer memory 12
The polygon id of the polygon is stored in (step S
46). Next, it is determined whether or not the processes of steps S43 to S46 have been completed for all the pixels included in the polygon (step S47).
Return to 43. If the processing has been completed, it is determined whether or not the processing of steps S42 to S47 has been completed for all polygons (step S48), and if not completed, the processing returns to step S42. If it is completed, it means that the hidden surface removal process is completed. Therefore, the element of the attribute vector is calculated for each pixel, and the color / luminance is calculated based on the calculated element (step S49), and this routine is completed. To do.

FIG. 7 is a flow chart for explaining the procedure of attribute value calculation, which corresponds to step S49 of FIG.
First, the y coordinate of the coordinates (x, y) on the screen indicating the position of the pixel to be processed is initialized to 1 (step S51). Next, the coordinates (x, y) on the screen
The x coordinate of the above is initialized to 1 (step S52).
Next, the polygon id of the coordinate (x, y) on the screen is read from the frame buffer memory 12 (step S53). Next, the attribute element ti specified by the polygon id
The function gi for calculation is selected (step S54).
Next, using the function gi and the coordinates (x, y) on the screen, the attribute element ti is calculated by the expression ti = gi (x, y) (step S55). Next, the calculated attribute element ti is stored in the position (x, y) of the attribute buffer memory 15 (step S56). These steps S54 to S56 are executed for all necessary attribute elements. Next, the x coordinate is incremented by 1 (step S57). Next, the value of the new x coordinate is compared with the maximum value xmax of the x coordinate on the screen (step S58). If the new x coordinate is less than or equal to Xmax, the process returns to step S53. The new x coordinate is Xma
If it is larger than x, the y coordinate is incremented by 1 (step S59). Next, the new y coordinate value is compared with the maximum y coordinate value Ymax on the screen (step S6).
0), if the new y coordinate is Ymax or less, step S5
Return to 2. If the new y coordinate is larger than Ymax, this routine ends. The frame buffer memory 12 stores the color / luminance calculated based on the attribute element.

As described above, since the attribute value is calculated only for the pixels corresponding to the image after the hidden surface removal processing, the invalid processing can be eliminated. Moreover, since a plurality of attribute elements are handled, a remarkable effect can be obtained for a high quality image as compared with the first embodiment. For example, the element (Nx, Ny, Nz) of the normal vector at each pixel is selected as the attribute element,
Suppose gi (x, y) is a linear expression for (x, y),
By obtaining the brightness of the pixel from the position and direction of the light source and the calculated normal vector, a higher quality (and therefore more realistic) phone shading algorithm (for example, "basic graphics") than the Gouraud shading algorithm is obtained. Satoshi Kawai, Shokodo, 1985
Year, pages 174-179). (Fourth Embodiment) In the third embodiment, the attribute value is calculated according to the procedure shown in FIG. 7, but the attribute value may be calculated according to the procedure shown in FIG. That is, first, the y coordinate of the coordinates (x, y) on the screen indicating the position of the pixel to be processed is initialized to 1 (step S61). Next, similarly, the x coordinate of the coordinates (x, y) on the screen is initialized to 1 (step S62). Next, the polygon id of the coordinate (x, y) on the screen is read from the frame buffer memory 12 (step S63). Next read polygon i
d is compared with the polygon id of the adjacent pixel (step S64), and when the read polygon id is equal to the polygon id of the adjacent pixel, a function g for calculating the attribute element ti
A value g ′ (x, y) of the partial differential component of i (x, y) in the x direction at pixel (x, y) is calculated (step S65). Next, using the attribute element values t′i and g′i (x, y) in the pixel (x−1, y) adjacent to the pixel (x, y), the attribute element ti of the pixel (x, y) is used. The expression ti = t'i + g'i
It is calculated by (x, y) (step S66). Step S
In 64, the polygon i read in step S63
If it is determined that d is not equal to the polygon id of the adjacent pixel, the function gi for calculating the value of the attribute element designated by the read polygon id is selected (step S6).
7). Next, using the function gi and the coordinates (x, y) on the screen, the attribute element ti is calculated by the expression ti = gi (x, y) (step S68). After performing the process of step S66 or step S68, the calculated attribute element t
i is stored in the position (x, y) of the attribute buffer memory 15 (step S69). The processes of steps S65 to S69 are performed for all necessary attribute elements. Then, the x coordinate is incremented by 1 (step S70). Next new x
The value of the coordinate is compared with the maximum value Xmax of the x coordinate on the screen (step S71), and the new x coordinate is Xmax.
If the following is true, the process returns to step S63. The new x coordinate is Xm
If it is larger than ax, the y coordinate is incremented by 1 (step S72). Next, the new y-coordinate value is compared with the maximum y-coordinate value Ymax on the screen (step S73). If the new y-coordinate is less than Ymax, the process returns to step S62. If the new y coordinate is larger than Ymax, this routine ends.

As described above, if the partial differential component g '(x, y) of the function g (x, y) in the x direction is used, the calculation becomes easy and the processing speed can be increased. The partial differential component of the function g (x, y) in the y direction may be used.

[0026]

As described above, according to the present invention, the value of the brightness I or the attribute element ti is directly calculated for each pixel on the screen after the hidden surface removal processing is performed on all the graphic data. Therefore, there is an excellent effect that the process for the part that does not contribute to the image generation of the final result is deleted and a higher quality image can be efficiently generated.

[Brief description of drawings]

FIG. 1 is a block diagram of a three-dimensional image generation device that realizes a three-dimensional image generation method according to a first embodiment of the present invention.

FIG. 2 is a flowchart illustrating a procedure of a three-dimensional image generation method according to the first embodiment of the present invention.

FIG. 3 is a flowchart illustrating a brightness calculation procedure adopted in the three-dimensional image generation method according to the first embodiment of the present invention.

FIG. 4 is a flowchart illustrating a brightness calculation procedure adopted in the three-dimensional image generation method according to the second embodiment of the present invention.

FIG. 5 is a block diagram of a three-dimensional image generation device that realizes a three-dimensional image generation method according to a third embodiment of the present invention.

FIG. 6 is a flowchart illustrating a procedure of a three-dimensional image generation method according to the third embodiment of the present invention.

FIG. 7 is a flowchart illustrating an attribute value calculation procedure adopted in the three-dimensional image generation method according to the third embodiment of the present invention.

FIG. 8 is a flowchart illustrating an attribute value calculation procedure adopted in the three-dimensional image generation method according to the fourth embodiment of the present invention.

FIG. 9 is a block diagram of a three-dimensional image generation device that realizes a conventional three-dimensional image generation method.

FIG. 10 is a flowchart illustrating a procedure of a conventional three-dimensional image generation method.

[Explanation of symbols]

1 processor 2 frame buffer memory 3 Z buffer memory 4 display device 11 processors 12 frame buffer memory 13 Z buffer memory 14 Display 15 attribute buffer memory

Claims (6)

[Claims] 1. A coordinate (x,
y) and the coordinate z representing the depth from the screen plane
In the three-dimensional image generation method for generating a three-dimensional image using three-dimensional graphic data defined by screen coordinates (x, y, z) consisting of and luminance I representing brightness, the screen coordinates A hidden surface erasing processing step based on a Z-buffer algorithm that determines the front and back of a figure for each pixel using the depth coordinate value z of the above, and selects the figure data closer to the viewpoint; And a luminance calculation step of calculating the luminance I for each pixel on the screen after the surface erasing processing step is completed. 2. The brightness I of a graphic can be uniquely determined by the (x, y) coordinate value of the screen coordinates, that is, can be defined as I = g (x, y) using a function g. 3. The three-dimensional structure according to claim 1, wherein in the calculating step, the brightness I is directly calculated from the position of the pixel, that is, the (x, y) coordinate value by using the formula I = g (x, y). Image generation method. 3. The brightness I of the graphic can be uniquely determined by the (x, y) coordinate value of the screen coordinates, that is, can be defined as I = g (x, y) using the function g, and immediately before. The luminance I ′ and the function g in the pixel for which the luminance value is calculated
Partial differential component g '(x,
y) and I = I ′ + g ′ (x, y) can be calculated, in the luminance calculation step, the figure of the pixel for which the luminance I should be calculated is the figure of the pixel for which the luminance I ′ was calculated immediately before. When the same, the brightness I is calculated by the above formula I = I '+ g' (x, y)
If the figure of the pixel for which the brightness I is to be calculated is different from the figure of the pixel for which the brightness value was calculated immediately before, the brightness I
The three-dimensional image generation method according to claim 1, wherein is calculated directly using the formula I = g (x, y). 4. Coordinates (x,
y) and the coordinate z representing the depth from the screen plane
An attribute vector (t1, consisting of a set of three-dimensional graphic data defined by screen coordinates (x, y, z) consisting of and attribute data for determining characteristics unique to the graphic such as color, brightness, and normal vector. In the three-dimensional image generation method for generating a stereoscopic image by using t2, ..., Tn) (n is an integer of 2 or more), a figure for each pixel using the depth coordinate value z of the screen coordinates. Before and after the hidden surface removal processing step based on the Z-buffer algorithm for selecting graphic data closer to the viewpoint, and after completion of the hidden surface removal processing step for all graphic data, attributes for each pixel on the screen A three-dimensional image generation method comprising: performing an attribute value calculation step of calculating a value of an attribute element ti (i is an integer from 1 to n) that is an element of a vector. 5. When the value of the attribute element ti can be uniquely determined by the (x, y) coordinate value of the screen coordinates, that is, when the function gi can be defined as ti = gi (x, y), In the attribute value calculation step, the value of the attribute element ti is directly calculated from the position of the pixel, that is, the (x, y) coordinate value, using the equation ti = gi (x, y). The three-dimensional image generation method according to item 4. 6. The value of the attribute element ti can be uniquely determined by the (x, y) coordinate value of the screen coordinates, that is, can be defined as ti = gi (x, y) using the function gi, and immediately before. The value of the attribute element t'i calculated in the above and the function g
Partial differential component g'i with respect to the x coordinate or the y coordinate of i
Using (x, y) and ti = t'i + g'i (x, y)
If the figure of the pixel of the attribute element ti whose value is to be calculated is the same as the figure of the pixel of the attribute element ti ′ whose value was calculated immediately before, the value of the attribute element ti is calculated by ti = t'i + g'i (x, y)
When the figure of the pixel of the attribute element ti whose value is to be calculated is different from the figure of the pixel of the attribute element ti ′ whose value was calculated immediately before, directly using the equation ti = gi (x, y) The three-dimensional image generation method according to claim 4, wherein the three-dimensional image generation method is performed as follows.