JPS5846026B2 - Hidden line cancellation method in display devices - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a hidden line elimination method for handling a three-dimensional space in a display device 1 whose image memory is a digital IC memory.

When displaying an object expressed in a three-dimensional coordinate system (x, y, z) three-dimensionally on a two-dimensional CRT screen, line drawings that display the boundaries and contours of the surface The three-dimensional objects curve like wirework, and even lines that would normally be hidden behind the real objects appear, making the diagram even more understandable.

To eliminate this, you can erase the invisible line from the diagram.

This operation is called hidden line elimination. Also, when trying to display a three-dimensional object by using only one surface instead of a line drawing (though conventionally such surface display has rarely been done), it is necessary to erase the object (which disappears due to hidden objects). In a broad sense, this is called hidden line elimination.

Various methods have been studied in the past in order to perform this hidden line elimination in a rational manner.

for example. There is a method to obtain hidden line elimination data using software known as the WARNOCKS and LOUTREL algorithms, but this type of software requires complicated data input and a huge amount of calculation. This requires a large load on the computer, and the processing speed is also unsatisfactory.

On the other hand, due to the recent development of IC memory technology1, digital IC memory has been used to develop image memory for monitor display (Vid.
io Frame Buffer), and it has become possible to perform high-resolution and high-speed display processing.

The present inventors assume a system in which the image memory is composed of digital ICs, and by utilizing the characteristics of this digital I'C image memory, it is possible to easily and quickly realize a variety of graphical displays that have been relatively difficult in the past. Various processing methods are being developed.

One of them is a technology called surface brightness generation method.
Based on this, the hidden line r white paper method of the present invention was developed.

First, an outline of a display device to which the present invention is applied will be explained with reference to FIG.

This device includes a display processor 1゜DDAD path 2, a memory system circuit 3, and an image memory 4 for monitor display.
, a CRT monitor circuit 5, etc., and also uses one or two image memories 6 and 7, which will be referred to as "Shirado-use" as will be described later.

The display processor 1 is a host computer (
This is where the display file given is expanded to obtain the necessary original data for generating straight lines, curves, or surfaces, including scaling, rotation, etc. (not shown).

The DDAD path 2 is a circuit that calculates image data based on the original data obtained by the processor 1, and this circuit outputs image data expressed in three dimensions (x, y, z). Output and this data is stored in the memory system (goods)
The data is written into the image memory 4, 6, and 71 via the circuit 3.

The image memory 4 is composed of a digital IC memory, and for example, assuming that the CRT screen is divided into 1 pixel of 1024 x 1024 (that is, 1 pixel is divided into 1 pixel on the CRT screen).
024 x 1.024 resolution), 1024 x 102 that corresponds two-dimensionally to each one-sided element array
The image memory 4 is composed of a plurality of memory planes each having an array structure of 4 memory cells.

The coordinate position (X
, the storage location specified by the Y address) each can store 1 bit of information, and multiple memory planes are arranged in parallel) This is used for each coordinate position; each can store multiple bits of information. This allows image data with display brightness information to be stored.

In the illustrated example, the image memory 4 is composed of four memory planes 4a to 4d, can write 4 bits of information at each coordinate position, and can express 16 levels of brightness.

Incidentally, the white clay image memory 6 and the image memory 7 are also exactly the same as the image memory 4 described above.

Now, in this case, when displaying a plane having a certain contour, closed loop data representing the positional interpolation of the contour line of the screen image at one point in the three-dimensional space is provided.

In response to this, a predetermined calculation is performed in the DDAD path 2, and as the calculation progresses i, the coordinate values (both x and y) on the X-Y plane of each point constituting the closed loop (both x and y (this 10-bit data)) are calculated. and the display brightness information (4-bit data, which is the Z-axis data of the three-dimensional coordinate system) of the point represented by the total X and Y standard values are sequentially processed at high speed in the DD open circuit 2. Output.

Incidentally, such calculation of image data (line segment data) by the DA circuit 21 is a known technique.

The memory system circuit 3 stores the X, Y
All standard values are X for image memory 4 (or 6, 7),
Supplying it as a Y address also inputs Z data as write data.

As a result, the image memory 4 stores the luminance data of each point corresponding to the X and Y coordinate values sequentially output from the DDA circuit 2.

When displaying the contents of the image memory 4 on the CRTI,
In the CRT monitor circuit 5, the raster scan of the CRT and the readout scan of the image memory 4 are made to correspond two-dimensionally and are synchronized, and the readout data from the image memory 4 is used as a luminance signal of the CRT. .

For example, when the above-mentioned closed loop data is written in the image memory 4 and is monitored by the CRT monitor circuit 5, a closed loop is drawn on the CRT screen by a line with continuous brightness changes.

This image is realized only by contour lines, and changes in brightness representing the positional relationship of the plane in the Z direction are also expressed only by the contour lines.

This is a common surface expression in the past.

Here, the above-mentioned surface brightness generation method 1, which is the basis of the present invention, will be explained.

In this method, when closed loop data as described above is given, and when a surface display command is given next to that closed loop, the closed loop is displayed as a continuous line defined by the brightness of each point on the closed loop. This is a signal processing method for generating such image data based on the closed-loop data (based on the above-mentioned closed-loop data) and writing it into the image memory 41, in order to display a screen without line drawings filled with bright lines having brightness changes. .

To explain in detail, the above-mentioned closed-loop data outputted from the DDA circuit 2 is first written into the image memory 6 for white clay production, and then, as described below, this image memory 6 is sequentially scanned in a fixed direction to write the closed-loop data. Brightness data is calculated while being read.

FIG. 2 schematically shows the state in which closed-loop data is written to the image memory 6.

However, although the closed loop 11 in the figure does not represent the luminance data, the image memory 6 stores the luminance data.

Image memory 6 has one X and Y axis (address) set as shown in the figure, and image memories 4 and 7 are exactly the same)
, the maximum and minimum of the X and Y addresses of the closed loop data written in this are XMAX, YMAX, and YMIX, respectively.
shall be.

For example, it is assumed that scanning for reading out this closed loop data on the image memory 6 is performed sequentially from the point (XMIN IYM IN+1) to the point (XMAX, YMAX 1) in the positive direction of the X-axis.

If the scanning range is limited to one in this way, two intersections with the closed loop l can always be detected for each scanning.

Although the explanation is complicated, a condition that satisfies this condition is naturally attached to the condition of the form of a closed loop l, and here we assume that the object to be regenerated is a closed loop that satisfies that condition. .

As mentioned above, two intersections with the closed loop l are detected for each scan of the image memory 6, and when scanning a certain scanning line Yi, one intersection point and one intersection point in the scanning direction are detected. The first point detected is called the first intersection point P1, and the next point detected is called the second intersection point Qi.

Also, the X and Y addresses of Pi are X (P l ) t
Y (P + ), and the luminance data of Pi is called Z (Pi
) (the same applies to the point Q i l).

Now, in the surface brightness generation method described above, in the closed loop data on the image memory 6, for example, 2
The luminance data of the two intersections Pi and Qi are Z (P
+ ) tZ(Q+), one monitor display image memory 4 (or 7) is used to calculate the point Pi -(,
X(Pi), Yi and point Qi-(X(Qi).

On the line connecting up to Yi)), from Z (Pi) to Z(Q
The luminance data is written by changing continuously by 1 at a uniform rate up to i).

Note that the point Pi in the image memory 41 as shown in 1 above,
The brightness data of Qi is stored in the image memory 61 as Z(P
It is most preferable to match exactly one point Pl) and Z(Qi)l, but the point Pl in the image memory 41
.

The luminance data of Qi is stored in the image memory 6) as z(pl
)tz(Q;) even if it differs by just one.

Therefore, data generation as described above is realized by, for example, the following control means 1.

(1) When reading and scanning the closed loop data written in the image memory 6, the scanning address signal is also applied to the image memory 4 for addressing, and the memory 4 is placed in a writing state.

When the line Yi of the image memory 6 is scanned, the intersections Pi and Qil are detected once, and X(Qi)X(Pi)-ΔXi corresponding to the distance between the two points is temporarily stored; Memorize the luminance data Z(Qi') of the second intersection Qi.

() → When scanning the next line Y i + 1; First, if the first intersection P i + 1 is detected, its luminance data Z (Pi + 1) is temporarily stored and 1, Z (
Qi) Calculate Z(Pi+1)−ΔZ1.

0 When scanning the above line Y i + 1, the first intersection P
In the scanning process from i+1 to the second intersection Qi+1, in synchronization with the scanning clock, from Z(Pi+1) to Z(Pi+1)+ at a rate of change corresponding to △Zl/△X i I.
Luminance data that continuously changes by 1 up to ΔZt is calculated and generated.

It is assumed that the write data to the luminance data 4 is generated sequentially at the same time as (E) and (E).

From the above, brightness data that continuously changes from Z(Pi+1) to Z(P i+1 )+Z i is written in the image memory 41 on a straight line connecting point Pi+1 to point Q i+1.

By sequentially performing this control from YMIN-1-1 to YMAX-1, two vertices (Y
The area of the image memory 4 corresponding to the entire area except for the line MIN and YMAX and the line YMIX+1 of the first scan line gradually changes the brightness to 1 on the closed loop.
It is filled in with luminance data connected to this area.

This is surface image data, and as mentioned above, this is 1 piece C.
When displayed via the RT monitor circuit 5, a surface with a contour represented by the above-mentioned closed loop and continuous brightness changes is depicted on the CRT screen.

In addition, as shown above, the two vertices of one closed loop and YMIN
Although data for the +1 line is missing, this is of no practical problem at all.

Further, by scanning YMIX+1 twice, it is easy to generate data for this one line.

Figure 3) shows the main part of the circuit for carrying out the above-mentioned controls (a) to (e).

In FIG. 1, Zout5 indicates the 4-bit luminance data successively output from the image memory 6, and Z+n4 indicates the luminance data to be written into the image memory 4.

As described above, the image memory 6 is sequentially scanned in areas where the closed loop data is written, and the scan addresses are stored in the image memory 41.

At this time, the 4-bit output Zout6 from the image memory 6 is logically summed for each digit in the OR gate G1, and the output is inputted to the flip-flop FFI.

This flip-flop FF is inverted at the rising edge of the output of OR'7"-1-G; in this way, the first intersection point is detected at each scan of the image memory 6. It is set when the second intersection point is detected, and is reset when the second intersection point is detected.

Therefore, the set output Q of the flip-flop is during the scanning period u 1 +1 from the first intersection to the second intersection.

The counter 10 is a set output Q of the flip-flop FF.
While G is 1'', the scanning clock CL is inputted through the AND gate G2 and is counted.

Reset output Q of flip-flop FF is '1'
When l rises, the count value of the counter 10 is transferred to the register 11.
When 1 is latched, 1 is latched, and then the counter 10 is cleared.

That is, for each scan, the counter 10
X(Pi) is counted and the value is latched into the register 11.

Also, what is the set output Q of the flip-flop FF? +11
+ When it rises by 1, the Zout output at that time
6, that is, Z(Qi), is latched by the register 131.

So now, the scanning of line Yi is finished and the next line Y
Assume that a scan of i + 1 is started.

At this time, the register 111 is X (P i )
(Qi), register 13 (here Z(Qi) is latched.

Then, after one scan of line Yi+1, the first intersection P
When i + 1 is detected, Z(Pi+1) is set in the register 12 and counter 141, and the subtracter 15 calculates △Z1-Z(Qi)-Z(Pi+1), and the value is expressed as a signed absolute value. The value conversion circuit 161 is supplied,
It is separated by 1 from the sign of: △Zi l.

The code signal becomes a signal that controls up-counting and down-counting of the counter 14, and also 1△Zi
l is read into one predetermined digit position of the register 18 via the selector 17.

The data X(Qi) Determine the digit position to be read.

The circuit section 22 surrounded by a dotted line, including the counter 14, the conversion circuit 16, etc., is a normal 1-axis vector generator that uses a DDA integrator as a constant multiplier. In the adder 201, the contents of the register 18 and the register 21 are added, and the contents are added to the register 211 again.
The carry signal generated from 0 becomes the count input for counting up or down for the count 14.

This operation is a general operation that has been used in vector generators for a long time, and will not be described in detail.

From the required 1 calculation of the vector generator 2-2, the output of the counter 14 is ( This is the preset Z(Pi+
From 1), the number of insects increases continuously at a uniform rate of change up to Z(Pi+1)+ΔZi, or becomes an adult.

The output of this counter 14 is derived via AND gates 03 to G6 to which the set output Q of the flip-flop FF is applied as a gate signal, and becomes the luminance data Zin4 to be written into the image memory 41.

Note that the above calculation is performed using the illustrated vector generator 22.
Not only that, but the same thing can be done with various types of vector generators.

By the way, the circuit in Fig. 3, that is, the one that generates the luminance data connecting 11, P i + 1 to Qi + 1 with the above (A) to (2), X (Pi), X (Qi), Z (Pi), Z(Qi).

Since the calculation is performed based on Z(Pi+1), the final value of luminance data obtained by this calculation is Z(Pi+1)+△Zi=Z(Pi+1)+4Z(Q
i)-Z(Pi+1)) may not match the luminance data Z(Qi+1) on the image memory 6.

However, even in that case, the difference between the two is so small as to pose no practical problem.

Because the above closed loop data (X.Y,,Z]IDD
It is calculated and given by A circuit 2, and each axis X,
The Y and Z data each show continuous changes, and Z(Q
This is because the difference between i) and Z (Q i + 1) is 1 bit or less.

For the same reason), the luminance data connecting Pi+1 to Qi+1 is generated].
There is almost no problem even if the calculation is performed as Zi=Z(Qi)Z(1)i).

Conversely, if line Yi+1 is scanned twice, P 1-
To generate luminance data connecting t-1 to Q i + 1, X(P i+1 ), X(Qi-t4
), Z(P i+1 ).

The calculation can be done based on Z(Qi+1), and the deviation is more accurate that way.

I will summarize the surface brightness generation method 1 explained above.

Closed loop data (X.

After writing data (Y, Z) into one image memory, when reading and scanning the closed loop data; the period from the time of detection of the first intersection point Pi to the detection of the second intersection point Qi is synchronized with the scanning. , Z(P i ) to Z(Qi), which is the gist of this surface luminance generation method.

Therefore, in this specification, generating the above-mentioned continuously changing luminance data Zin based on the closed-loop luminance data Zout that is read out and scanned is called surface luminance calculation, and an example of this is a circuit as shown in FIG. is called a surface brightness calculation circuit.

It is assumed that the memory control circuit 3 shown in FIG. 1 is provided with a surface brightness calculation circuit.

Now, the first aspect of the hidden line elimination method, which is the gist of the present invention, will be explained.

FIG. 4 (As shown in this figure, a case where a rectangular parallelepiped is displayed in a perspective view will be explained as an example (note that it is not meant to be represented by a line drawing as shown in the figure).

All target values (X, Y, Z) of the eight vertices a to h shown in the figure are given as original data of this rectangular parallelepiped.

In the first invention, the rectangular parallelepiped is understood as a set of the following six planes.

○Boundaries ab, bd, dc, ca (closed loop 11)
Surface S1 surrounded by 0 boundary lines ab, bf, fe, ea (closed loop 12) S2 0 Surface surrounded by boundary lines ac, cg, ge, ea (closed loop 13) S3 0 boundary Surface S4 surrounded by lines ba, dh, hf, fb (closed loop 14) Surface S5 surrounded by boundary lines cd, dh, hg, gc (closed loop 15) 0 Border lines ef, fg, hg, ge (closed loop 16) Then, in the display processor 1 and the stages before it, each plane 81 to S6 is treated as closed loop data (x, y, z) representing the closed loop 1146 that is its outline, and from the closed loop data, The above-mentioned surface brightness calculation circuit 1 is responsible for generating the data (1).

Next, a processing procedure for displaying the rectangular parallelepiped shown in FIG. 4 in surface representation with hidden line elimination will be explained in order.

(1) First, any one of the closed loops 11 to 16 above
For example, the closed loop 14 generates closed loop data from the DDA circuit 2 and writes it into the image memory 6.

(2) Next, the image memory 6 and the image memory 4 are scanned synchronously, and the closed loop data 14 read out from the image memory 6.
1, perform the above-mentioned surface brightness calculation based on the image memory 41, and write the calculated brightness data to the image memory 41.
Surface data S4 representing the image S4 on the image memory 4 is generated.

Note that I in FIG. 5 represents surface data S4 on the image memory 4.

However, the figure does not show the changes in brightness of the surface data.

I don't do it.

(3) Next, the image memory 6 is cleared, and any other one of the above closed loops, such as closed loop 1, is cleared.
1 is generated from the DDA circuit 2, and written into the image memory 6.

(4) Next, the image memory 6 and the image memory 4 are scanned synchronously, and the results are sequentially obtained by the above decay brightness calculation based on the closed loop data 111 read from the image memory 6 S11
The problem is to write the luminance data to the image memory 4 (in other words, the calculated luminance data is not written unconditionally, but instead the data already written in the image memory 4 (i.e. surface luminance data) is written. S4) and compares it with the calculated output surface brightness data S1,
If the surface brightness data S1 is large), this is the case when the surface brightness data S1 is large.
If the surface image data S4 is large, one is written to the image memory 4 (the data written here is left as is).

The control circuit for this purpose is shown in block form as shown in Figure 6.
This will happen.

In other words, the image memories 6 and 4 are scanned synchronously, and the closed-loop luminance data Zoui6 read out from the image memory 6 is input to the surface brightness calculation circuit 301, and from this circuit 30, the surfaces S and l are input. The stuck luminance data is sequentially output in synchronization with scanning, and this becomes new write data Zin4 to the image memory 4.

On the other hand, in the one-sided image memory 4, the data written therein is first read out in the first half of the control cycle for one address, and the data Zout4 and the new data Zin4 are compared in the comparator circuit 31. , when Zin4>Zout4, a write command signal WR is issued from the comparator circuit 31, and from this, in the latter half of the control cycle, new data Zin4 is written to the address at that time.

In addition, if Zln4≦Zout4, the image memory 41
Since the above-mentioned write operation is not performed on this address, data Zout4 remains at that address.

(Incidentally, ff11tll in the above (2) corresponds to 1 in the image memory 4) in this (4) in a state where not a single word has been written.

) By the control in 4 above, in the image memory 4, at a point on the surface S4 and a point not included in the surface S1, the surface S4
brightness data is written, and also at the point of surface s1 and surface S
For points not included in 4, the brightness data of surface S1 is written,
Furthermore, at the point where the surface S4 and the surface s1 overlap, the brightness data of the larger one, that is, the surface S in front in the Z direction
1 brightness data is written.

Therefore, in this case, as shown in FIG. 5H, the entire surface data is written for the surface S1, and the data for the portion overlapping with the surface S1 is erased for the surface S4.

This is the result of hidden line elimination in the relationship between surface S1 and surface S2.

And yet another closed loop, for example closed loop 1
Regarding 3, if you perform the control described in 3 and 4, the image memory 4
As shown in Figure 5 ■, the soil has three sides S1. In the relationship between S4 and S5, surface image data from which hidden lines have been removed is generated.

Similarly, when the control for the six closed loops 11-16 is completed, image data is completed in the image memory 4 as a surface representation of a rectangular parallelepiped with hidden lines completely removed, as shown in Figure 5. It is.

When this is outputted to CI' (T) by the CRT monitor circuit 5, the hidden lines are erased and a rectangular parallelepiped made up of surfaces with luminance changes corresponding to the Z-axis coordinate value is displayed.

Note that the above explanation is based on three surfaces S1. S2. Although we have taken as an example a simple rectangular parallelepiped with poles in which S3 is completely displayed and the other three surfaces S4, S, , S6 are completely erased, the hidden line cancellation display method of the present invention can be used to It is clear from the principle of the present invention explained above that three-dimensional objects can also be handled in the same way.

Next, an invention for expressing a three-dimensional object with shading will be explained.

In the hidden line erasure display method described above, the brightness of each point on the surface of a three-dimensional object that is not erased is determined by the Z-axis coordinate value of that point.

On the other hand, in the second hidden line elimination display method that I am going to explain, light is irradiated from a certain direction on the three-dimensional object (to be displayed), and a certain surface on the three-dimensional object is %1 brighter. This is intended to express a state in which a surface becomes particularly dark, creating a shadow.

For example, in Figure 5
Assuming a state in which light is irradiated from the degree direction, S3 is brighter than the brightness defined by the Z-axis coordinate value of each point on the surface,
This method attempts to express the surface S2 darker than the luminance defined by the Z-axis coordinate value of each point on the surface, and a method for generating such image data will be described below.

FIG. 7 shows a circuit block diagram for the second hidden line elimination display method.

The relationship between the image memories 6 and 7 in the same figure is exactly the same as the relationship between the image memories 6 and 4 in FIG. (1) When steps (1) to (4) are executed, the brightness is displayed on the image memory 7, as shown in Figure 5. Image data consisting of data is generated, but in this invention,
In parallel with the generation of the above-mentioned image data on the image memory 7, image data with the above-mentioned shading is generated on the image memory 4.

That is, in terms of the circuit, the brightness data output from the surface brightness calculation circuit 30 becomes manual data Zin7 to the image memory 7, and is also inputted to an arbitrary offset value Zo in the adder 321. are added to and subtracted from each other to become the input data Zin4 to the image memory 4.

Further, from the comparison 1 between the data Zout7 read from the image memory 7 and the input data Zin7, the comparison circuit 3
The write command signal WR output from 1 is supplied not only to the image memonor but also to the image memory 4.

Further, the three image memos IJ 4, 6, and 7 are scanned synchronously.

Therefore, by setting the offset value Zo to zero, the upper 1d
When the control described in l) to (4) is performed, as is clear from the explanation so far, the image memory 7 and the image memory 4 are
Exactly the same image data is generated in each case.

However, if the closed-loop data 12 has already been written in the image memory 61, and the offset value Zo is set to an appropriate negative value during scanning to generate surface data S2 while reading it, the image The brightness absorption data of the surface S2 written on the memory 41 has a value smaller by the offset value a than the luminance data (written on the image memory 71) defined by the Z-axis coordinate value.

In this way, brightness data that does not faithfully reflect the Z-axis coordinate value is written in the image memory 41, but this data is not provided as comparison data for subsequent hidden line elimination. Since the comparison calculation for removing the hidden line 1 is performed using the data written in the image memory 71, no problem occurs.

Similarly, when scanning to generate the surface data S3 based on the closed loop data 13 (here, if the offset value Zo is set to an appropriate positive value b1, the brightness data of the surface S2 written in the image memory 41 will be , the Z-axis coordinate value is larger than the value of the run-up by the offset value.

As described above, image data with the above-mentioned shading added is generated.

The hidden line elimination display method 1 of the present invention 1 has been explained in detail above, but this method 1 is based on the digital IC.
By using an image memory that corresponds two-dimensionally to the display screen, calculation of the surface brightness based on the closed loop data 1, and generation of hidden line removed data based on the calculated surface brightness data 1, Alternatively, surface data with added shading can be generated at high speed using very simple logic.

Moreover, the input data required to display such a three-dimensional object is mainly closed-loop data that specifies the outline of each surface that makes up the three-dimensional object, and this does not include hidden line removal. The amount of data is almost the same as when displaying a simple line drawing.

Furthermore, when displaying with one shading added as described above, it is sufficient to provide certain closed-loop data (accompanied by the above-mentioned offset value).

In this way, the configuration of the input Y-Y becomes simple.

That is, according to the present invention, data for hidden line elimination, which was conventionally obtained as a result of enormous calculations using a large-sized computer, can be obtained with a simple configuration, although some hardware, mainly an image memory, is required. It can be generated extremely quickly.

This is an extremely significant effect in increasing the versatility and practicality of the display.

[Brief explanation of drawings]

FIG. 1 is a black diagram showing a schematic configuration of a graphic display device to which the present invention is applied, FIG. 2 is an explanatory diagram of a surface brightness calculation method, and FIG. 3 is a block diagram showing essential parts of an example of a surface brightness calculation circuit. FIG. 4 is an explanatory diagram of a rectangular parallelepiped to be displayed in the hidden line elimination display method. FIG. 5 is an explanatory diagram of the process of generating hidden line removed data, FIG. 6 is a block diagram showing the main parts of a circuit that realizes the hidden line removed display method of the first invention, and FIG. FIG. 2 is a block diagram showing the main parts of a circuit that implements the hidden line elimination display method of the invention. 1...Display processor 2...
...DDA circuit, 3...Memory control circuit, 4.
...Image memory for monitor display, 5...C
RT monitor circuit, 6, 7... Image memory for Shiradonari, 30... Surface brightness calculation circuit, 31...
- Comparison circuit, 32... Adder.

Claims (1)

[Scope of Claims] 1. A method for eliminating hidden lines from a three-dimensional object composed of a plurality of planes in a display device using an image memory composed of a digital IC memory, wherein each of the planes is (a) writes only the closed-loop data to the first image memory; (b) writes the closed-loop data to the first image memory; The written closed loop data is sequentially scanned one time in a fixed direction, and the read luminance data is inputted into a surface brightness calculation circuit 1. From this surface brightness calculation circuit 1, in synchronization with the scanning, the closed loop data is read out. (c) In parallel with (b) above, scan the second image memory in synchronization with the first image memory, and output from the surface brightness calculation circuit. Compare the brightness data with the brightness data written in advance in the second image memory, and if the brightness data output from the surface brightness calculation circuit is large, write that data into the second image memory. By performing the above control of (a), (b) + (C) on each of the closed loop data, one hidden line is erased from the second image memory, and the above three expressed as
A hidden line elimination display method in a graphic display device that generates image data of a dimensional object. 2 A method for eliminating hidden lines from a three-dimensional object composed of a plurality of planes in a display device 1 using an image memory composed of a digital IC memory, in which each of the planes is (a) writes only the closed-loop data into the first image memory; (b) writes the closed-loop data into the first image memory; (The closed loop data written here is read out by sequentially scanning in a certain direction, and the read luminance data is input to the surface brightness calculation circuit, and from this surface brightness calculation circuit 1, the data in the closed loop is read out. (C) In parallel with (b) above, scan the second image memory in synchronization with the first image memory, and calculate the luminance data output from the surface brightness calculation circuit. and the brightness data written in advance in the second image memory, and if the brightness data output from the surface brightness calculation circuit is large, write that data into the second image memory, (d ) above, 1ilffb) l Parallel to (C),
The third image memory is scanned in synchronization with the second image memorization, and an arbitrary value is added or subtracted from the brightness data to be written to the second image memory among the brightness data output from the surface brightness calculation circuit. , write to the third image memory, perform the control of the above (a) t (b), (c), and (d) for each of the closed loop data 1. From this, hidden lines are erased on the third image memory. Hidden line portrait display method in a display device that generates image data of the three-dimensional object, which is represented by a surface with arbitrary shading added thereto.