JPS60243772A - Method for painting out inside of loop on display device - Google Patents

Abstract

PURPOSE:To paint out efficiently the inside of a loop with simple constitution by checking the vertical change of border points every detection of a new border point and painting-out points on the basis of the checked result. CONSTITUTION:When a painting-out request is applied to an optional point (Xb, Ya) in an area surrounded by a loop L1, the intersected point (Xa, Yb) between a horizontal line passing the point and the closed curve L1 are detected as a starting point at first, then while moving the loop L1 in the clockwise direction Ar1 e.g., the horizontal line is successively painted out. When the painter is returned to the starting point (Xa, Ya) after one cycle, the painting-out in the area of the loop L1 will be completed.

Description

DETAILED DESCRIPTION OF THE INVENTION <Industrial Application Field> The present invention provides a rask scan type display device in which closed curve data representing the outline of an image is written in a display memory.
The present invention relates to a filling method that can efficiently fill in the closed curve area with a simple hardware configuration.

<Prior art> In a rusk scan type display device, for example, as shown in FIG.
When a filling request expressed by is given, points within the closed curve area are filled in as shown by . in FIG. 2, using this as a starting point.

For example, in 1982, A
ddison-wesley pub.

Published by J. D. FoIey.

A, D, Dam common chopsticks”Fundamental so
fInteractive Compute-er
A method such as that described in "Graphics" P, 448-451 is known.

The first method shown in this document is that when a filling request is given to an arbitrary point within an area surrounded by a closed curve, the requested pixel is centered around the requested pixel and is applied in four directions (left, right,
(adjacent above, below), or 8 directions (left, right, above, below, diagonally adjacent)
Check for more adjacent pixels according to certain rules,
The pixels to be filled in are sequentially identified.

FIG. 3 is an explanatory diagram showing a state in which filling is performed using such a method. When a filling request, represented by S, is given within a closed curve region represented by 0, processing according to the recursive definition is performed repeatedly starting at 8 points, and pixels to be filled are sequentially identified. In the case of this specific example, it moves downward from the starting point S by 1 and 2, hits the lower boundary point, bends to the left, and moves upward by 3.4.
...Turns back to 7, hits the upper boundary point, bends to the left, and then goes to the lower side with 8.9. Filling is performed in a fixed order, such as A, B, and so on.

The second method disclosed in this document is a method in which, when a filling request is given within an area surrounded by the closed curve, pixels in the horizontal direction adjacent to the area are sequentially filled. In this case, after filling is completed, the upper and lower horizontal lines adjacent to the filled horizontal line are scanned to find the next portion to be filled. The data obtained by this is accumulated in the stack,
The next part to be filled in is selected from among these data, and this is then filled in.

4 to 6 are explanatory diagrams showing the state of rows (rows) filled in using this method. In FIG. 54, when a filling request represented by S is given, Adjacent horizontal pixels are filled in sequentially, as shown by . After this, the upper and lower horizontal lines adjacent to the filled horizontal line are scanned, and the data regarding these is accumulated in the stack. In Figure 4,
1°2 is a representative representation of the stack of these data. In this specific example, pixel group 2 is selected as the next horizontal portion to be filled in, and this portion is filled in as shown in Figure Fi5.

After this, as in the case of FIG. 4, the L side and lower horizontal lines adjacent to the horizontal line that has been filled in are scanned to identify the area to be filled next, and as shown in FIG. Pixel group 2 in the figure is filled in.

However, in the case of the first method, although the respective procedures are simple, it is necessary to repeatedly perform processing according to the recursive definition for each pixel, which takes time. there were. Furthermore, when memory space is limited, the stack may overflow.

In addition, in the second method, it is necessary to scan all pixels within the area of the closed curve twice, which is time-consuming. In addition, in order to find the next part to be filled, the horizontal line that has been filled is Adjacent horizontal lines must be scanned, and the data m from this scan changes significantly depending on the form of the figure created by the closed curve.
In order to be able to correctly handle even complex shapes, it was necessary to provide a sufficiently large memory.

Purpose> The purpose of the present invention is to represent the outline of an image! The object of the present invention is to realize a filling method that can efficiently perform filling in the area of the closed curve with a simple hardware configuration in a rask scan type display device in which J111 curve data is written in a display memory in advance.

<Configuration of the Invention> The configuration of the present invention is such that in a rask scan type display device in which closed curve data representing the outline of an image is written, when a request to fill in any point within the closed curve area is given, this point is Move in the horizontal direction to find a point where it intersects with the closed curve, use this as a starting point, scan the pixels adjacent to this point clockwise or counterclockwise to find the point of intersection with the closed curve, and then scan this point anew. The operation of scanning the same boundary point as in the previous case and detecting a new boundary point is repeated along the closed curve to the starting point, and each time the boundary point is detected, the operation of detecting a new boundary point is repeated. When the vertical change tendency is in a specific direction, the area is filled in from this boundary point horizontally to the point where it intersects with the closed curve, and the area surrounded by the closed curve is filled in.

<Example> First, the principle of the filling method according to the present invention will be explained in the seventh section.
The explanation will be given according to the drawings and FIG. The horizontal and vertical coordinate positions are represented by Xn and Yn. Any point (X, b,
When a request to fill in Ya) is given, first find the intersection (xa, Ya) between the horizontal line passing through this point and the closed curve L1, and use this point as the starting point. From this point, the closed curve L1
For example, while moving in the direction Ar1 around the clock, the horizontal line is sequentially filled with X's. - Go around and start point (Xa, Y
When the process returns to F3), filling in the area of the closed curve L1 is completed.

The process of moving along the closed curve L1 in the clockwise direction Ar1 will be explained with reference to FIG. FIG. 8 is an enlarged view of the closed curve L1 shown in FIG. 7, with the starting point (Xa, Ya) represented by a black dot as the center, and pixels adjacent to it from eight directions are shown. There is.

Among these, the diagonally shaded points are pixels forming the closed curve L1, and this closed curve is written in the display memory in advance.

Once the starting point (Xa, Ya) is determined, the addresses of eight pixels adjacent to this point in the counterclockwise direction Ar2 are checked. The point that first intersects with the closed curve L1, that is, (Xa
-1', Ya+1) is specified as a new boundary point. Next, using this as a new starting point, scanning is performed in the same manner as in the previous case to detect new boundary points in the history. By repeating these operations, the starting point (>(a, Y
It is possible to trace the closed curve up to a) above 1.

Further, each time the boundary point is detected, filling is performed from the previous boundary point to the point horizontally intersecting the closed curve 1-1. In this case, in order to prevent the same horizontal line from being filled twice, if the vertical change tendency of the boundary point is in a specific direction, for example, Yn of the new boundary point
When the value of increases compared to that of the previous boundary point1
Fill in the horizontal line.

By the way, the shape of the closed curve L1 is not necessarily a simple shape as shown in FIG. There may also be cases of complex closed curves. Examples of the present invention in such a case will be described below. FIG. 9 is an enlarged view of a portion of the complex closed curve L2, and black and diagonally shaded points represent pixels forming this closed curve. In the following explanation, the process of identifying the starting point on the closed curve L2, that is, a request to fill in an arbitrary point within the closed curve L2, and the intersection of the horizontal line passing through this point with the closed curve L2 is aimed. , a description of the process used as the starting point is omitted. Now, from the position of the sunspot, arrow Ar3
It is assumed that the vehicle travels along a closed curve L2 in the direction shown in FIG.

FIG. 10 shows a control circuit for implementing the filling method of the present invention, and FIG. 11 shows a flowchart for explaining its operation. According to these, the filling state in FIG. 9 will be explained.

In FIG. 10, ALLI is an arithmetic unit, AB and DB
is a data bus, and 013 is an output bus. X has an X value of R
EAD, register to be WRITE, Y is Y value is REA
D, register to be WRITE, Prx I is closed curve L
2, the boundary one before the current position (closed curve L2
The leftmost X471i of > is the register to be read and written, prx r is the rightmost X value of the boundary immediately before the current position is READ.

The register to be WRITE, CX l, is the leftmost X value of the boundary of the current position is READ, WRI Tl: the register to be written, Cxr4. The rightmost X value of the boundary of the tm location is RjE
AD, WRI TE register, Pry, y value of the previous boundary from the current position is READ.

The register to be written to, Cy is the boundary y of the current position
The register CNTR in which values are read and written is a counter. MEM is a memory in which graphic information related to closing song 11L2 is stored, and CLOG is a control for executing various logical operations shown in Figure 11.
It's logic.

Now, at the position of the sunspot, the leftmost xi (Cx l)
is α, the rightmost X value (CXr) is β, and yl (Gy) is γ. As is clear from Fig. 9, the leftmost X value (prxl) and the rightmost X value (prxr) of the boundary immediately before the current position are β, and yili of the previous boundary
(Pry) is γ-1.

The process of checking the addresses of pixels adjacent to the two black points in a counterclockwise direction, detecting the intersection with the closed curve IIL2, and filling them out will be described with reference to the flowchart of FIG. 11. In step (1), prxr+
1 <=β+1) and written to register X.

In step (2), Cy-,1<=r-1> is determined and written to register Y. These steps determine the starting point of the scanned pixel.

In step (3), Cxr-prxr(-0) is determined and stored in counter CNTR. Step (4
), the value of counter CN 1' R is OJX-1
m is determined. In this example, it is 0, so proceed to step (5), where ×(-β+1>, Y(=γ-
It is determined whether 1) is on the boundary. Since X and Y are not on the boundary, the process moves to step (6), where the value of counter CNTR minus 1 is stored as a new counter value, and in step (7), X+1 is stored as the new value in register X.
written to.

Thereafter, the process returns to step (4), but in this example, the value of the counter CNTR becomes -1, the loop is exited, and the process moves to step (8). In addition, in step (5),
, Y are on the boundary, the process moves to the steps of filling B and below and initializing.

In step (8), Cxr+1 (=β+1) is determined, this is written to register X, and in step (9), CV+1
(=γ+1) is written to register Y, and in step (10) cx r−Cx I +2
(=3), and this is stored in the counter CNT'R. These steps and subsequent steps (11) to (14) correspond to the process of scanning the black point (β, γ) at the right end to the left from the diagonally upper pixel (β+1.γ+1).

Note that the pixel to the right of the black point (β, γ) is skipped and moved to the pixel diagonally above the right, but the pixel to the right was previously detected at the boundary by scanning to detect the previous boundary point. It has been confirmed that this is not present (see the initialization steps below B below), so it is skipped.

In step (11), the value of counter CNTR is 0.
It will be judged if it is more than that. In this example, it is 3, so proceed to the next step (12) and here, ×(−β+1>,
It is determined whether Y (-γ+1) is on the boundary. Since X and Y at this time are not on the boundary, the process moves to step (13), and the value obtained by subtracting 1 from the value of counter CNTR is stored as a new counter value, and in step (14),
1 is written to register X as a new value.

After this, the process returns to step (11), but in this example, this loop is repeated three times, and the process moves to steps 13 and below on the fourth time.

The steps below A are the pixels diagonally to the upper right of the black dot at the right end (
β+1. From γ+1) to the diagonally upper left pixel of the black point at the edge of No1 (
The steps taken when the boundary is not found by scanning up to α-1, γ+1) are not shown.

If there is a boundary at the position indicated by the dotted line in FIG. 9, scanning must be performed further counterclockwise from the black point at the left end. Steps <15) and (16) are the scanning starting point (α
-1, γ-1). Note that the starting point skips the pixel to the left of the leftmost black dot and starts from the diagonally lower left pixel (α-1, γ-1) of the black dot. This is because the pixel to the left of the black dot is , this is because MtAl was determined not to be on the boundary during the scanning process to detect the previous boundary point.

At step (17), prXl-QX+ is stored, and at step (18) it is determined whether the value of the counter CNTR is 0 or more. In the next step (19), it is determined whether X and Y are on the boundary, and if they are not on the boundary, step (20)
), the value of the counter CNTR minus 1 is stored as a new counter value, and in step (21), X+1 is written to the register Xk as the new value.

Thereafter, the process returns to step (18), but if the position of the dotted pixel is on the boundary as in this example, this loop is exited at the -th time and the process moves to steps B and subsequent steps. Furthermore, the step (
If 18) is NO, it means that the closed curve L2 is discontinuous.

In the steps below B, step (22) is a filling subroutine, and the y of the current boundary (black point position) is
la (Cy), the y value of the previous boundary (Pry), and the y (direct (Y)) of the boundary detected by the current scan are compared, and when Pr y < CV < Y, that is, the current When the change tendency of the boundary points is increasing, filling is performed from the position (Cx r+ 1 , Cy,) next to the rightmost black point to the opposite boundary to the right.

Steps (23) to (32) are initialization steps performed prior to scanning to detect the next boundary point.

In steps (23) and (24), the rightmost X value of the new boundary is determined. If the right edge is hit, step (
In step 25), the rightmost X value (CXr) of the current boundary is written to the register prXr relating to the rightmost X value of the previous boundary, and in step (26), X-1 (in step (23), X+1→ ) is written to register Qxr.

In steps (27>, (28>), the X value of the left end of the new boundary is found. If the left end is found, step (
29), the leftmost X value of the current boundary < CX +
> -゛]Register Pr for the leftmost X value of the previous boundary
x l, and in step (30), write X+1(
In step (27), X-1→×.

) is written to register Cxl.

In steps (31) and (32), the y value is updated. That is, in step (31), the y value (cy> of the current boundary) is written to the register Pry related to ylIfi of the previous boundary, and in step (32),
The new boundary y value is written to register Cy 8Q to the current boundary y value.

Thereafter, the process returns to step (1) and scanning is performed to detect a new boundary point.

By repeating these operations, the closed curve L2 is traced to the starting point 'g' indicated by the black dot, and the area surrounded by this closed curve is filled in. still,
In the embodiment of the present invention, a new boundary point is detected by scanning counterclockwise around the boundary point.
The present invention is not limited to this, and the boundary point may be detected by scanning around 51 hours.

Effects> According to the filling method of the present invention, the memory contents for scanning around a boundary point are coordinate data immediately before, presently, and immediately after the boundary point, which is smaller than the conventional method, and the memory scale is small. can be reduced accordingly. Furthermore, according to the method of the present invention, a small number of pixels adjacent to a boundary point are scanned, a new boundary point is found, and a straight line is drawn (because
As shown in the first conventional method, the filling process is repeated according to the recursive definition for all pixels within a closed curve, or as in the second conventional method, filling is performed for each horizontal line. , the time required for filling can be greatly reduced compared to a system in which adjacent horizontal lines are not scanned one by one in preparation for the next filling. In addition, in the method of the present invention, the vertical change trend of the border point is checked every time a new border point is detected, e.g., only when the current border point trend is increasing. Fill in the horizontal line, and the same horizontal line is 2r! 1. Since it is not filled in, time can be further saved.

[Brief explanation of the drawing]

1 to 6 are block diagrams of a road using the conventional paint-over method, and FIG. 11 is a 71-coach yard diagram for explaining the operation of the method of the present invention. 1. L2...Closed curve, △l-LJ... Arithmetic device,
X. Y, Prx l, Prxr, Cx l, Cxr, C'/
. Pry...Register, MEM...Memory for storing graphic information, CLOG...Control logic. Figure 1 Figure 3 0000000000000000000000000 0 0 0 ooooooooooooooooooooo. ooooooooooooooooooooo. 5th prisoner ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo Figure 7 Figure 8 Hayabusa Figure 9

Claims (1)

[Scope of Claims] In a last metascan type display device in which closed curve data representing the outline of an image is written, (a) when a request to fill in any point within the closed curve area is given, this point is (b) Using this as a starting point, pixels adjacent to this point are scanned clockwise or counterclockwise to find a point of intersection with the closed curve; (C)! (d) when the tendency of change of the boundary point is in a specific direction;
A method for filling inside a closed curve in a display device, which comprises a process of filling in horizontally from this boundary point to a point where it intersects with the closed curve.