JPH04184688A - Boundary point generator for plane polygon - Google Patents

Abstract

PURPOSE:To prevent a back triangle from being plotted by selecting and painting out an integer lattice point on or nearest to the ridge line of a plane triangle by an adder means. CONSTITUTION:A ridge line interpolation result is supplied to an accumulative addition value register 1 via a multiplexer 6, and a gradient value in a direction of paint-out is supplied to a gradient value register 2, and initialization is completed. Thence, a constant value after the round-up of decimeal in a first constant value register 4 is selected, and is sent to an adder 3 together with the ridge line interpolation result, and a result is fed back to the register 1 via the multiplexer 6. A value corresponding to the integer lattice point inside the plane triangle on or nearest to the ridge line can be held with the register 1 by performing such processing. After that, the gradient value of the register 2 is selected by a multiplexer 7, and paint-out is performed sequentially. The value of a completion point of paint-out is selected from a second constant value register 5 which holds a numetric value rounded up and subtracted by one, therefore, no picture element is shared on the boundary of the triangle.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention relates to a plane polygon boundary generation device, and more specifically, the present invention relates to a plane polygon boundary generation device. Relating to a device for generating.

<Prior Art> Conventionally, when drawing a figure, a figure drawing device has been provided which divides a figure of an arbitrary shape into a plurality of triangular polygons and performs a filling process on each of the obtained triangular polygons. ing.

In this figure drawing device, all figures to be filled in are triangular polygons, and triangular polygons are always flat, so before the drawing process is performed by the figure drawing device,
A configuration is adopted in which the gradient in the filling direction and the gradient in the edge direction are obtained, and the edge interpolation calculation and the filling interpolation calculation are pipelined in this order based on the vertex data and gradient data of the triangular polygon obtained in advance.

Specifically, the contents of the cumulative addition value register (71) that holds vertex data or cumulative addition data and the gradient register (72) that holds the gradient are supplied to the adder (73), and the addition result is used as the cumulative addition value. At the same time, the content of the cumulative addition value register (71) is output as the edge cumulative addition result. Then, the contents of the cumulative addition value register (74) that holds the edge line cumulative addition result or cumulative addition data as an initial value and the slope register (75) that holds the gradient are supplied to the adder (76), and the addition result is It feeds back to the cumulative addition value register (74), and also feeds back to the cumulative addition value register (74).
) is output as the coloring cumulative addition result. It should be noted that the result of the cumulative addition of 8 colors is output as an integer value in order to fit into an integer lattice for display.

<Problems to be Solved by the Invention> In the conventional graphic drawing device described above, when drawing adjacent planar polygons, the drawing pixels at the boundary become the same. However, depending on the drawing order, the back polygons may end up being drawn, resulting in a disadvantage in that the drawing quality of the figure as a whole deteriorates.

Additionally, when performing pixel-by-pixel operations such as raster operations, the pixel-by-pixel operation is performed multiple times on the same pixel, and there is no guarantee that the desired result will be obtained. be.

<Object of the invention> This invention was made in view of the above problems,
It is an object of the present invention to provide a planar polygon boundary generation device that can generate boundaries between adjacent planar polygons using points included inside the respective planar polygons as boundaries.

Means for Solving the Problems> To achieve the above object, a boundary generation device according to a first aspect of the present invention includes a cumulative addition value holding means for storing edge interpolation data of a plane polygon or a filling cumulative addition value; a gradient value holding means for holding a gradient value in the direction; a first constant value holding means for holding a constant value for rounding up the decimal part; and a second constant value holding means for holding a constant value for rounding up the decimal part and subtracting by 1. A numerical value holding means, a selection means for selecting the contents of the slope value holding means, the first constant value holding means and the second constant value holding means, and adding the contents of the cumulative addition value holding means and the contents selected by the selection means. and an addition means for supplying the cumulative addition value again to the cumulative addition value holding means.

However, the gradient value holding means, the first constant value holding means and the second
Preferably, the constant value holding means is a multiport register file or a multiport memory.

The boundary generation device of the third aspect of the invention includes cumulative addition value holding means for holding edge line interpolation data of plane polygons or filling cumulative addition values, gradient value holding means for holding gradient values in the filling direction, and gradient value holding means for holding gradient values in the filling direction; a first constant value holding means for holding a constant value, a second constant value holding means for holding a constant value for rounding up the decimal part and subtracting by 1, a cumulative addition value holding means, a first constant value holding means, and It includes a selection means for selecting the contents of the second constant value holding means, and an addition means for adding the contents of the gradient value holding means and the contents selected by the selection means and supplying the sum again to the cumulative added value holding means. .

Effects> With the boundary generator of the first five aspects of the invention, when obtaining a filled pixel on an integer grid by performing an interpolation operation between two points obtained by edge interpolation, for the leftmost filled pixel, The contents of the first constant value holding means are selected by the selection means and added by the addition means to obtain filled pixels on the edge line or on the right side of the edge line, that is, inside the plane polygon to be drawn. Thereafter, by selecting the contents of the gradient value holding means using the selection means and performing cumulative addition, pixels to be filled in can be sequentially obtained. For the rightmost filled pixel, the selection means selects the contents of the second constant value holding means and the addition means performs the addition, so that the pixel on the edge line or to the left of the edge line, that is, the plane polygon to be drawn, is You can get the filled pixels inside.

As is clear from the above description, since only the filled pixels inside the corresponding plane polygon can be generated for all adjacent plane polygons, the inconvenience of sharing a boundary can be solved. In other words, in the conventional boundary generation method, since the edges are shared, if a filled pixel on the edge line that is common to adjacent plane polygons is inside the plane polygon for one plane polygon, it will be inside the plane polygon for the other plane polygon. Since it was always outside the plane polygon,
Depending on the drawing order, the back polygon may be displayed at the matching part between the front polygon and the back polygon, which may be an inconvenience.Also, in pixel operations, etc., where calculations are performed on a pixel-by-pixel basis, the same filled pixel may be displayed multiple times. However, in the present invention, such a problem can be reliably solved.

With the boundary generator of the second invention, it is only necessary to select and read out the relevant contents from the multiport register file or multiport memory, which simplifies the configuration. effect can be achieved.

In the boundary generation device of the third aspect of the invention, the content of the first constant value holding means is selected by the selection means when supplying the edge line interpolation data of the plane polygon to the cumulative added value holding means as an initial value through the adding means. Since the addition is performed by the adding means, the boundary on the fill drawing start side can be generated at high speed, and the overall drawing speed can be increased.

<Example> Hereinafter, embodiments will be described in detail with reference to the accompanying drawings showing examples.

FIG. 1 is a block diagram showing an embodiment of a graphic drawing device incorporating the boundary generation device of the present invention, which includes a cumulative addition value register (1) that holds the edge interpolation result as an initial value;
The gradient value is cumulatively added to the contents of the gradient value register (2) that holds the gradient value in the filling direction of the plane polygon, and the cumulative addition value register (1), and the addition result is stored in the cumulative addition value register (1). Adder (3) to add and constant value “0000, FFFF (HEX
)' and a constant value "FFFF, F" for rounding up the decimal part and subtracting by 1.
FFF (HEX)” is stored in the second constant value register (
5), a multiplexer (6) that selects the edge interpolation result or addition result and supplies it to the cumulative addition value register (1), the contents of the slope value register (2), and the first constant value register (
4) or the contents of the second constant value register (5) and supplies it to the adder (3).
It has

The operation of the fill-in interpolation device having the above configuration is as follows. In the following description, filling interpolation calculations are performed from the left edge of the plane polygon toward the right edge.

First, the edge line interpolation result is supplied to the cumulative addition value register (1) via the multiplexer (6), and the gradient value in the filling direction is supplied to the gradient value register (2), thereby completing the initial setting.

Next, the content of the first constant value register (4) is selected by the multiplexer (7) and is supplied to the adder (3) together with the edge interpolation result. The addition result can be obtained by rounding up the decimal part, and this addition result is fed back to the cumulative addition value register (1) via the multiplexer (6). That is, by this process, a value whose integer part is an integer grid point on the edge line or inside the planar polygon closest to the edge line is held in the cumulative addition value register (1).

Thereafter, the contents of the gradient value register (2) are selected by the multiplexer (7), and interpolation results can be obtained sequentially in the same manner as in the conventional device.

Then, when obtaining the value of the filling end point, the contents of the second constant value register (5) are selected by the multiplexer (7) and supplied to the adder (3) together with the cumulative addition value, so that the edge interpolation result is Unless is a value on an integer lattice, the addition result can be obtained by rounding up the decimal part and subtracting 1. That is, by this process, a value whose integer part is an integer grid point on the edge line or inside the planar polygon closest to the edge line is held in the cumulative addition value register (1).

The contents of the cumulative addition value register (1) are output as they are as the interpolation results in the filling direction, and are supplied to an image memory (not shown) with the decimal part truncated. Therefore, regardless of the state of the plane polygon, the value on the grid point within the plane polygon can be output as the interpolation result in the filling direction, and the pixels on the grid point are shared at the boundary between adjacent plane polygons, which is an inconvenience. can be resolved.

Figure 2 is a block diagram showing the configuration of an interpolation calculation unit that performs interpolation calculations of depth values (hereinafter referred to as z values), and includes a multiport register that holds initial values and gradient values.
file (11), a bidirectional shift register (12) that shifts read data from the multiport register file (11), and an edge cumulative addition result register (13) that holds the cumulative addition result during edge interpolation. )and,
A filling cumulative addition result register (14) that holds the cumulative addition result during interpolation in the filling direction, a multiplexer (15) that selects the contents of the bidirectional shift φ register (12) or "0° data," and the cumulative addition result described above. A multiplexer (1B) that selects the contents of the register or "O° data, an adder (17) that performs addition based on the data selected by the multiplexer (15) and the data selected by the multiplexer (16), and the addition result. Pipeline register (18) that temporarily holds
It has Note that the addition result is supplied to one of the cumulative addition result registers.

The operation of the interpolation calculation unit configured as shown in Fig. 2 is explained in Figs. 3 and 4.
This will be explained below with reference to the timing chart shown in the figure.

(1) When not correcting the z value However, as shown in Figure 3 (
The case of the triangular polygon shown in G) will be explained.

In this case, the initial value z00 is first read out from the multi-board register file (11) as shown in FIG. 3(B), and then the bidirectional shift register is read out as shown in FIG. 3(C). The initial value z00 is output as is from (12). In this case, since "0' data is selected by the multiplexer (16), the initial value zoO is supplied to both cumulative addition result registers, and the initial value z00 is also supplied to the pipeline register (18). , are output at the next clock timing (the same IN
(D) (See (E) and (F)).

After that, as shown in FIG.
In addition to continuing to output the gradient value d x, the bidirectional shift register (12) also outputs the gradient value d
z/dx continues to be output, and as shown in FIG.
Continues to output 0. Then, by selecting the contents of the cumulative addition value register (14) for filling by the multiplexer (16) and performing cumulative addition, the binary values z01. z 02゜z 03. z 0
4. z 05 can be obtained sequentially (see the same figure).

(E) (see (F)).

After the interpolation calculation in the filling direction is completed, that is, the binary z
After obtaining 05, the bidirectional shift register (12)
The gradient value dz02/dy02 in the ridgeline direction is output from the multi-board register file (month) through , and the content zDO of the cumulative addition value register for ridgeline (15) is outputted, so it can be added by the adder (17). As a result, a binary initial value zlO for the next filling direction is obtained. Then, the obtained initial value zlO and the above gradient value d z
/ d By performing the cumulative addition operation based on x, the next filling direction binary value z 11. z 12・
... can be performed sequentially.

Note that FIG. 3(A) shows the clock signal output.

Thereafter, by repeating the above series of operations, binary values for the entire range of the plane polygon can be obtained.

(Il) When correcting the z value However, the calculation of zM from the point PA to PB shown in FIG. 4 i) will be explained.

In this case, both cumulative addition value registers (13) (14)
and pipeline register (18) to Figure 4 (F
)(G) ()I), the last interpolated values Ez5. In the state where Ez55 is being output, as shown in FIG. 4(B), the multi-board register file (11) first outputs the gradient value dzo1/dy01 in the ridge direction, and then the gradient value dz in the filling direction. / d x continues to be output. The gradient value dzo1/dyo1 is output as is from the bidirectional shift register (12), as shown in FIG. (17) (same figure (
D) (See (E)).

Therefore, by performing addition by the adder (17), a new zf[Ez6 on the ridgeline is obtained, which is supplied to both cumulative addition value registers (13) and (14).
After that, the above z ii E z 6 is output from both cumulative addition value registers (13) and (14) ((F) (G) in the same figure).
)reference).

After that, first the multi-board register file (
11), the bidirectional shift register (12) outputs the gradient value dz/dx, and then the bidirectional shift register (12) is sequentially shifted to the right one bit at a time. gradient value dz/2dx,
dz/4dx, dz/8dx.

d z / 16 d x is output (see (C) in the same figure).

Then, the gradient value d z / 2 d x is calculated for the value Ez6 held in the cumulative addition value register for filling.
.

dz/4dx, dz/8dx
, d z / 16 d x can be sequentially cumulatively added to achieve correction of the Z value. For example, if the decimal part of the X coordinate value is a binary number and is 1010, the value is “1”.
Select the slope value corresponding to the ° digit, and if the value is 0°, select 0° data instead of the slope value (see figure (D)), and the final Binary A corrected to
z3 can be obtained. In addition, the intermediate two values are AzO-
Ez6 +dz/2dx, Azl-AzO1Az2 =
Azl +dz/8dx, Az3-Az2.

After that, by cumulatively adding the gradient values d z / d x using Az3 as the initial value of the binary values in the filling direction, the binary values z 61 . z 132... can be obtained.

Note that FIG. 4(A) shows the clock signal output.

Thereafter, the drawing of the figure can be achieved by repeating the above process as many times as necessary.

<Embodiment 2> FIG. 5 is a block diagram showing another embodiment of a graphic drawing device incorporating the boundary generation device of the present invention. The difference from the embodiment of FIG. 1 is that the gradient value register (2) and the gradient value in place of the multiplexer (7), and the constant value “000
0. FFFF (HEX)“.

The only difference is that a multi-vote register file or multi-vote memory (9) that holds "FFFF, FFFF CHEX)" and can be selectively read is used.

Therefore, in this embodiment, the same effect as in the embodiment of FIG. 1 can be achieved by simply selecting and reading out the gradient value or constant value from the multi-vote register file or multi-vote memory (9). .

<Embodiment 3> FIG. 6 is a block diagram showing still another embodiment of a graphic drawing device incorporating the boundary generation device of the present invention.
In addition to the configuration shown in the figure, a constant value “0” for rounding up the decimal part is added.
000. The first constant value register (19) holds FFFF (HEX)°, and the second constant value register (20) holds the constant value “FFFF, FFFF (HEX)” for rounding up the decimal part and subtracting by 1. The contents of the edge line cumulative addition result register (13), the contents of the filling cumulative addition result register (14), and the first constant value register (1
9) and the contents of the second constant value register (20) are selected by a multiplexer (I6).

Therefore, in the case of this embodiment as well, it is possible to generate a boundary located inside a planar polygon as in the above embodiment, and also to achieve correction of the z value.

In this embodiment, instead of moving the initial value to the cumulative addition value register and then adding a constant value, it is possible to add the constant value to the initial value and supply it to the cumulative addition value register. The drawing speed can be improved somewhat.

It should be noted that the present invention is not limited to the above-described embodiment; for example, it is possible to use a unidirectional shift register instead of the bidirectional shift register (12).
Various design changes can be made without departing from the gist of the invention.

<Effects of the Invention> As described above, the first invention has the unique effect of eliminating the inconvenience of sharing boundaries, since it is possible to generate only the filled pixels inside the corresponding plane polygon for all adjacent plane polygons. play.

The second invention has the unique effect of simplifying the configuration and achieving the same effect as the first invention.

The third aspect of the invention has the unique effect of being able to quickly generate a boundary on the fill drawing start side and increasing the overall drawing speed.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of a graphic drawing device incorporating the boundary generation device of the present invention, FIG. 2 is a block diagram showing the configuration of an interpolation calculation unit that performs depth value interpolation calculation, and FIG. 4 is a timing chart explaining the operation of the interpolation calculation unit shown in FIG. 2, FIG. 5 is a block diagram showing another embodiment of a graphic drawing device incorporating the boundary generation device of the present invention, and FIG. The figure is a block diagram showing still another embodiment of a graphic drawing device incorporating the boundary generation device of the present invention, and FIG. 7 is a block diagram showing the configuration of a conventional graphic drawing device. (1)... Cumulative addition value register, (2)... Gradient value register, (3)... Adder, (4) (19)...
First constant value register, (5) (20)...Second constant value register, (7)...Multiplexer, (11)...Multi-vote register file, (
14)... Cumulative addition result register for coloring, (1
5)...Multiplexer, (17)...Adder, Fig. 5

Claims (1)

[Claims] 1. Cumulative addition value holding means (1) for holding edge interpolation data of plane polygons or filling cumulative addition values, gradient value holding means (2) for holding gradient values in the filling direction, and decimal A first constant value holding means (4) for holding a constant value for rounding up the decimal part, a second constant value holding means (5) for holding a constant value for rounding up the decimal part and subtracting by 1, and a slope value holding means means (2), selection means (7) for selecting the contents of the first constant value holding means (4) and the second constant value holding means (5);
), the contents of the cumulative addition value holding means (1) and the selection means (7
2.) Adding means (3) for adding the contents selected by (1) to the accumulated added value holding means (1). 2. Gradient value holding means (2), first constant value holding means (4)
2. The planar polygon boundary generation device according to claim 1, wherein the second constant value holding means (5) is a multiport register file or a multiport memory. 3. Cumulative addition value holding means (14) for holding edge interpolation data of plane polygons or filling cumulative addition values, and gradient value holding means (11) for holding gradient values in the filling direction.
), a first constant value holding means (19) for holding a constant value for rounding up the decimal part, and a second constant value holding means (20) for holding a constant value for rounding up the decimal part and subtracting by 1.
and selection means (16) for selecting the contents of the cumulative addition value holding means (14), the first constant value holding means (19) and the second constant value holding means (20), and the slope value holding means (11). An apparatus for generating boundary points of plane polygons, comprising an adding means (17) for adding the content and the content selected by the selection means (16) and supplying the sum again to the cumulative addition value holding means (14).