JPS63228274A - Clipping system in multiwindow - Google Patents

Abstract

PURPOSE:To attain a high speed processing by inspecting whether a graphic in question enters a window before a division to a rectangular area or not before a graphic clipping processing for every rectangular area and excluding the graphic which does not enter as one which is not an object. CONSTITUTION:At the time of clipping segments (graphic) L1-L3 on the window 14 (rectangular area), the clipping processing is carried out on all the window 14 including a part con hidden by the window 12 (whether segments L1-L3 enter the 14 or not is checked) and the segment (graphic) L3 which does not enter the window 14 is eliminated. Thereby, it is not necessary to execute the clipping on a wasteful rectangular area, so that an efficient and high speed clipping processing can be carried out.

Description

DETAILED DESCRIPTION OF THE INVENTION [Summary] In a screen control method that manages a non-rectangular window as a sum of a plurality of rectangular regions, clipping processing is performed on the entire window before clipping processing is performed on each rectangular region of a line segment. Also, this clipping may cause a break between two adjacent rectangular areas, but this is filled in by the process of finding the intersection with the virtual line of the Fll interval.

[Industrial application field]

The present invention relates to a clipping processing method for graphics spanning multiple rectangular areas on a multi-window display screen.

[Conventional technology]

There is a multi-window display screen that allows multiple windows to be set overlappingly on the display screen.

FIG. 8 shows an example of this, where 10 is a display screen;
12, 14, and 16 are windows set to partially overlap within this screen 10, with 12 at the top, 14 below 12 and above 16, and 16 above 10 and below 14.

When drawing a line segment (shape) L+ in a certain window on such a multi-window screen, it is necessary to check whether the line segment is not completely or partially hidden by the window above (higher priority). . Since this process is rather troublesome, there is a method of dividing the window and managing it as shown in Fig. 9 (Japanese Patent Laid-Open No. 59-102284). In this method, the first window 12 has the highest priority, so it is managed as is. Since the window 14 is lower than 12, it is divided into two parts by a line hidden by 12 (in this example, a horizontal line, which is a straight line along the lower side of 12), 14a and 14.
Make two rectangles b. Furthermore, since the window 16 is below 14, it is divided into two by a line hidden by 14 (in this example, it is also a horizontal line, a straight line along the lower side of 14) to form two rectangles 16a and 16b. Screen 10 also has windows 12 and 14.
A straight line along the top and bottom sides of ゜16, and the rest is window 12
．． 14.16 is divided by straight lines along the left and right sides to form <10a, 10b, . . . 10h as shown in the figure.

When windows (these are rectangular) overlap, the lower windows become non-rectangular, but in the above method, all are divided into rectangular areas by the above-mentioned division, and the windows are managed in the state of this rectangular area.

The work of drawing the line segment L1 in a manner that manages the entire display screen as a collection of rectangular areas is performed by dividing the rectangular areas 14a and 14 into
Draw a line segment across b (this will be a task).

The line segment L2 is drawn across the rectangular areas 14a, 12, and 14b. Give priority to each rectangular area,
If 12 has the highest priority and 14a and 14b have the next priority, line segment L1 can be drawn directly between 14a and 14b.
The line segment L2 can be drawn to 14a and 14b excluding 12 parts, and the problem of hiding/appearing can be easily solved.

[Problem that the invention seeks to solve]

The base of the display screen is memory, and each of the rectangular areas is defined by a memory address. Since the addresses indicating the boundaries of each area are not the same but differ by 1, there are gaps at the boundaries of each area. This gap is one dot interval in the horizontal direction and one horizontal scanning line interval in the vertical direction, but in this case, both are assumed to be one dot interval. This type of display screen is represented by a collection of pixels (dots), which are not continuous but quantized in units of dots (figures on the screen are represented by a collection of points on a square grid with dot spacing).

FIG. 10 shows this situation, where a collection of points P on a square grid represents a line segment.

In the above method, the work of drawing a line segment across the rectangular areas 14a and 14b involves drawing a line segment P2P4 in the area 14a (in detail, each grid point from point P2 to P4 is displayed; the same applies hereinafter), and The task is to draw a line segment PIP3 in the area 14b. Point P1. Since P2 is the start and end point, it is usually a point on a square grid, but point P3 is calculated as the intersection of the lower side of area 14b and the line segment, and point P4 is calculated as the intersection of the upper side of area 14a and the line segment. Fractions may appear, but
The points are rounded to form points on a square grid.

In this way, the points on the boundary line of each area P: 1, P4 are the points on the square grid closest to the intersection of the boundary line of each area and the line segment.
When calculating, as shown in Figure 10, there are no display points between the image boundary lines (in this example, three white circles are missing),
On the display screen, the line segment appears to be broken at this portion (this tendency is more pronounced as the angle between the line segment and the boundary line becomes smaller). In order to display the same as the others and make this part appear continuous, it is necessary to add three white circles.

In addition, the work of drawing line segments on the screen involves finding the intersection of the line segment and the boundary line of each rectangular area, dividing the line segment up to that point (clipping), and drawing the line segment. but,
This is a time-consuming process, as it involves determining whether or not there is an intersection between the boundary line of each rectangular area and the line segment, and if so, determining the coordinates of the intersection point. Some rectangular areas are unrelated to (separate from) line segments, but
In the conventional method, the presence/absence of an intersection is checked for such items as well.

The present invention aims to improve this point and exclude line segments that are found to be irrelevant from being subjected to clipping processing. Further, in the present invention, line segments may be interrupted between the boundary lines, but this is attempted to be filled in and made continuous.

[Means for solving problems]

As shown in FIG. 1, in the present invention, when clipping the line segments (figures) L+ to L3 for (the rectangular area of) the window 14, the clipping process is performed for the entire window 14 including the portion hidden by the window 12 (
Line segments L1, L2. Check whether L3 falls within 14) and fill all (line segments (figures) that do not fit) into window 14. In this example, L3 is excluded (step 1). Through this process, part or all of the line segment (shape that does not fit) is excluded (step 1). Only the figures L1 and L2 that are considered to be the same are subject to clipping processing for each divided rectangle (step 2).

In this way, the figure L2, which is inside the window but is not displayed because it is in the shadow area of another window, remains in the above process and becomes the target of the clipping process for each divided area (this is based on the above-mentioned clipping degree). ).

For breaks in line segments at adjacent boundary lines, do the following: That is, as shown in FIG. 2, the adjacent boundaries BLI, BL
, an imaginary boundary line IL is drawn at a half dot interval d/2 between the lines 2 and 2, and the intersection point Pa between the line segment and the line IL is calculated. If there is a lattice point on the P3 side of the intersection Pa on the boundary line BL+ (there is one P5 in this example), that is also adopted as a point representing a line segment, and the intersection P on the boundary line BL2.

If there is a lattice point closer to P4 (in this example, there is one lattice point P6), that is also adopted as a point representing a line segment.

[Effect]

In this way, there is no need to perform clipping on useless rectangular areas, so that efficient and high-speed ripping processing can be performed.

Further, no dots are missing in adjacent boundary line portions, and no breaks occur in line segments that span both rectangular areas and intersect the boundaries of these areas at small angles.

〔Example〕

To determine whether a line segment falls into a window or not, use Cohe
It is preferable to use the n-Sutherland algorithm.

As shown in FIG. 3, the screen (the entire area) is divided into nine parts by extending the top, bottom, left, and right sides of the window, and a 4-bit code (4-bit code) is assigned to each partial area as shown in the figure.

The meaning of each bit is as follows (that is, the first bit (rightmost bit) is 1 means that the partial area is to the left of the left edge x1 of the window, and the second bit is 1. This means that the partial area is on the right edge of the window
To the right of That's what it means. By giving 4-bit codes in this way, you can easily determine whether a line segment with start and end points in these subregions falls within the window by taking the AND of the 4-bit codes of the subregions with start and end points. can be determined. For example, line segment L1 does not fit into the window, but the logical product of the start and end points of this line segment is (1001) and
(1010) = (1000), and since it is not 0, it may be assumed that it does not fit into the window. Also, line segment L2
The logical product of the start and end points is (0001) and (001
0)=(0000), which is O, so it enters the window. Also, the logical product of the start and end points of line segment L3 is (0
101) and (0110) = (0100), and since it is not 0, it does not pass through the window. Starting and ending points are 0
Some of the items in 001 and 1000 fit into the window, while others do not, but the logical product is 0. Therefore, some line segments with a logical product of 0 fit in the line segment, while others do not fit in the line segment, but
If the logical product is not O, it may be said that it is not included, and this is to be excluded.

In order to prevent discontinuities occurring in adjacent boundary line portions, the method shown in FIG. 2 is used, but the intersection point Pa between the virtual boundary line IL and the line segment, etc. can be determined as internal division points.

That is, as shown in Fig. 4, the point X that divides the line segment between the start and end points X1 and X2 into m; n is X = (n x + + mx 2 ) / (m + n
) −”---(1). Pa, PQ, P
If this relationship is applied to a, m = n, so the fl) formula is X = (x + + x 2)/2, and x of Pa,
The y coordinate value is the average value of the x and y coordinate values of Pa and Pa. PI, P2 to Pa.

To find Pa, first find m and n in equation (1), and then calculate equation (11).

Clipping processing is performed when a line segment is outside the window (more specifically, a rectangular area) as shown in FIG.
This is done except when the line segment is within the window as shown in figure bl. Since both the beginning and end are oooo, this can be detected.

Specifically, there are usually a large number of line segments since they are part of a figure, and those line segments to be clipped or not are selected in the manner described above.

The clipping procedure will be explained with reference to FIG. 5 (C1, (d)).

In this example, it is assumed that a line segment intersects the left edge of the window 14a at yL, and intersects the virtual boundary line IL at Po (the y coordinate is yV2). td) is a partially enlarged view of tc>. If the intersection yL is not at a grid point, it is quantized and yL' becomes the display point. When the display point is determined by intersection calculation, yL' is obtained,
If this ends, the line segment may be interrupted at the boundary line. To eliminate this, Pa-yx,
(Both y coordinates) = 8 is given to yL. y is this yL+
It is point ε.

Let dy be the y-coordinate difference between the start and end points of a line segment, and dx be the x-coordinate difference, then dx: dy-1/2: ε +'+ 5= 1, ay 2 dx (2) Also) 'L' ) 'L, = α. Based on the magnitude relationship between 6 and α, 7%, y%' becomes as shown in FIG. 6 (0) to (d).

FIG. 6(0) shows the case where ε=α=0, in which case the line segment is horizontal and the intersection with the window boundary line BL is on the grid point. Therefore'! L=Vr,'=V%=y'A'. The display point is yL'.

FIG. 6(b) shows the case where ε<α, in which case the quantization point yL' of yL and the quantization point y' of 7% become display points. y'
The quantization point of A is found on the yL side when viewed from y'A, but not on the opposite side (if it is found, the line segment will go too far at the boundary line and a whisker will appear).

Figure 6(C) shows the case of ε=α, in which case yL'-y
%'=y%, and yL' becomes the display point.

Figure 6(d) shows the case of 8>α, in which case the quantization point yL' of yL and the quantization point yz' of yVa, and yz' and yL
', the lattice point yV2'' (not limited to one, Iy%'
There are only yL' 1 and 'd 1) as display points.

Although only the display points on the boundary line of one window have been described in FIG. 6, the display points on the boundary line of the other window adjacent to this window are also determined and added in the same manner.

FIG. 7 shows the clipping processing procedure of the present invention.

(0) indicates a processing procedure for drawing a line segment within the window; if the line segment is outside the window, this line segment is not targeted and is excluded as having no drawing area. If all line segments are within the window, draw all the line segments in this window. If part of the line segment is outside the window, it will be clipped, and only the line segment inside the window will be drawn in this window. fc) indicates a starting point clipping process, which finds the intersections of the line segment with the left, right, bottom, and top sides of the window, and if there is one, the starting point of the line segment ends there.

If two intersections are found, the other is the end point. (d+ indicates the missing correction process of the boundary line part, which calculates the intersection between the line segment and the window boundary line (in this example, the left side), and also calculates the intersection point with the virtual boundary line, and from these, the display point is decide.

〔Effect of the invention〕

As explained above, according to the present invention, in a multi-window type display device that divides a window into rectangular areas, it is possible to prevent line segments (figures) from discontinuing at the boundaries of rectangular areas, and to Since the object is checked and the clipping process is not performed on line segments that are not the target, efficiency is increased and processing speed can be increased.

[Brief explanation of the drawing]

Figures 1 and 2 are illustrations of the clipping method of the present invention, Figure 3 is an illustration of the Cohen-Sutherland algorithm, Figure 4 is an illustration of the internal division method, Figure 5 is an illustration of clipping, Figure 6 is an explanatory diagram of the missing part correction procedure, Figure 7 is a flowchart of the clipping process, Figure 8 is an explanatory diagram of multi-window, Figure 9 is an explanatory diagram of the rectangular division method, and Figure 10 is a diagram of what occurs in the conventional method. It is an explanatory diagram of a problem.

Claims (2)

[Claims] (1) Multiple windows (12,
14) can be set partially overlapping, and the window area that becomes non-rectangular due to the shadow of another window is divided and set as the sum of multiple rectangular areas (14a, 14b,...), and multiple In a graphic clipping method in a screen control system that manages a window as a collection of rectangular areas, before the graphic clipping process for each rectangular area, the relevant graphic (L_1, L_
2,...) is included, and those that do not fit are excluded as out of scope. (2) Multiple windows (12,
14) can be set partially overlapping, and the window area that becomes non-rectangular due to the shadow of another window is divided and set as the sum of multiple rectangular areas (14a, 14b,...), and multiple In a graphic clipping method in a screen control system that manages a window as a collection of rectangular areas, before the graphic clipping process for each rectangular area, the relevant graphic (L_1, L_
2, ...) will be included, those that do not fit will be excluded from the target, and only those that will fit will be subject to clipping, and each opposing boundary line (BL_1, BL_2) of the adjacent rectangular area will be
), draw a virtual boundary line (IL) at the center of the figure, find the intersection point (P_0) between this virtual boundary line and the figure, and find the grid points between this point of intersection and the intersection points (P_3, P_4) between the figure and the boundary line. (P
_5, P_6) is added as a display point on the boundary line.