FI77180B - FOERFARANDE OCH ANORDNING FOER REGISTRERING AV SKRIVTECKEN. - Google Patents

Description

1 77180

Method And device for drawing characters

The invention relates to a method for plotting characters point by point and by image line, in particular by drawing characters in raster form on the basis of character data stored on a drawing medium, retrieving character data of characters to be drawn and converted into control commands for a drawing element, and a device for implementing the method.

An electronic light emitting device in which the drawing element is an electron beam tube is already known from U.S. Pat. No. 3,305Θ41. The characters required for the Ladon task are stored as letter information in the letter data memory. The text to be loaded is converted in the stack processor into text data, which are stacking instructions for the light stacker. While typesetting is in progress, the text information sequentially calls the letter information from the character characters needed to draw the characters, and the read letter information is converted into a video signal that provides control commands to the electron beam tube. Each character is drawn on the screen of the cathode ray tube as a plurality of line rasters consisting of adjacent vertical image lines extending in the direction of lines of text, the characters being formed from their black and white portions, respectively, by the electron beam brightness / dark setting.

On the electron beam screen, the drawn characters are illuminated by optics on the drawing medium.

In these known photomultipliers, the characters of the individual letters must be stored for all the font sizes present, so that the capacity of the character memory is very large.

To reduce the capacity of the character memory, it is common to store characters in the form of encoded letter information 2 77180. There are various coding methods for this. Point coding of characters is known, for example, from U.S. Pat. No. 3,305,841, durability coding of parts of an image line from U.S. Pat. No. 3,480,943, and encoding of border lines of characters from DE-29 19 013.

DE-24 22 464 (U.S. Pat. No. Re 30,679) already discloses a method for characterizing an edge line of a character and converting the edge-coded letter information into control instructions for an electronic beam of an electron beam for drawing or storing characters. This border coding describes the characters as border pairs in the XY coordinate system, and changes the x coordinates in equal length steps. Each pair of edge lines is defined by the y-coordinate of the starting point and the angle of inclination of the edge line, whereby the angle of inclination is expressed as changes in the y-coordinates of successive x-coordinate steps. The curvature of the outline of a letter is thus given as a series of varying angles of inclination. The characters are again formed from adjacent image lines on the electron beam tube running vertically against lines of text, with each character being drawn at a given point in the line of text to be stacked and all the characters in the lines of text sequentially. Control commands for switching the electron beam on and off are then generated by defining the respective intersections of the edge line pairs with the displayed image line, whereby the electron beam is always switched on between the upper and the associated lower contour.

The described method for generating control commands takes quite a lot of computation time, so a high display speed is not achieved. Although the use of an electron beam tube as a display element achieves a typographically high quality of photocomposition, it is disadvantageous that the usable picture tube area of the electron beam tube and thus the drawing size is limited. Today, it is hoped that entire pages of a newspaper could be stacked in one step. In order to stack such a newspaper page 3 77180 in an electron beam photoconductor, a transfer must be made between the individual sub-exposures of the size of the usable screen area and the drawing medium. This in turn requires a great deal of mechanical use, leads to positioning errors and slows down the recording speed.

EP-A-0096O79 already discloses a method and an apparatus for drawing entire newspaper pages in dots and in line lines from letter code encoded in a display grid from a sketch plan according to a sketch plan. In this case, a drawing element, for example a mosaic drawing element, is guided in pixels over the entire width of the entire newspaper page, so that in each case the complete image lines of the entire newspaper page are drawn. Each image line then consists of sub-image lines which, according to the sketch plan, belong to different text blocks or lines of text. The encoded letter information needed to draw individual sub-lines is retrieved from the letter information memory and decoded into control instructions. The control instructions for the individual sub-lines are arranged in a sequence corresponding to the sequence of the individual sub-lines within the entire picture lines and are stored in the screen raster for pixel-by-pixel memory so that the contents of the memory reproduce the entire newspaper page data by pixel. During the drawing of the entire newspaper page, the individual control commands are then given pixels and line by line in synchronization with the movement of the drawing element over the drawing medium, and applied to the drawing element.

In order to save memory capacity, it is already known from EP-A-0096079 to illuminate entire newspaper pages in successive strips, the height of which comprises several image lines and the width of which corresponds to the width of the entire newspaper page. For band-like drawing, only the coded letter information belonging to the individual positions of the band on the newspaper page is retrieved, control instructions are formed from them for a number of pixels corresponding to the band height, and stored in a so-called window memory with a capacity corresponding to only one pixel in 77180. Before moving the band to a new location on the newspaper page, the control instructions are transferred line by frame from the window memory to a buffer memory of equal capacity, and read from the buffer memory for controlling the drawing element while the freed window memory is refilled.

However, this deposit principle has the disadvantage that each time a band is transferred, a transfer time must be reserved for control instructions, which limits the drawing speed. The presented method works properly only when the height of the window memory corresponds to the height of the largest character to be drawn. However, this condition has the disadvantage that the required window memory height must always be reassigned according to letter instructions when entering text data, and memory control must be programmed accordingly, and that the maximum memory capacity must be selected so that the height condition is also met for the largest characters, e.g. newspaper page headlines. Another disadvantage of the described method is that storing control commands in the window memory is particularly complicated and time-consuming when the character to be stored is not completely in one lane (EP-A-0096079) or when two consecutive characters on one line of text have different base line heights (EP-A-00). ; Figure 14). Namely, in these cases, the control instructions for the first character must always be written first from the parts of the window memory to the buffer memory before the control instructions for the next character can be placed in these parts. The publication does not provide any advice on what to do when drawing italic characters and translated lines of text. It is already known per se to use a deflectable laser beam as a drawing element for drawing such newspaper pages.

SUMMARY OF THE INVENTION It is an object of the invention to provide a method and apparatus for pixel-by-pixel and line-by-line drawing, in particular lettering, by means of which characters are edge-coded, and by

5 771 80 This object is solved with respect to the method as claimed in claim 1 and with respect to the device as claimed in claim 25. Other claims provide advantageous further developments.

In particular, a high drawing speed and good drawing quality are achieved with the method according to the invention. With the given edge bar coding, it becomes particularly simple to convert the coded character data into control commands for all letter variations without having to store the character data for the individual variants separately. Moreover, the formation takes place without distortion and with high resolution.

The invention will be described in more detail below with reference to Figures 1-13. They set out:

Figure 1 shows the basic structure of the device for carrying out the method; Figure 2 is a graphical representation for explaining the exposure window and the search window; Figure 3 is a graphical representation of a letter rectangle; Figure 4 is an implementation example of a video signal generator; Figure 5 is a graphical representation of the interpolation; Figure 6 is a connection point diagram; Figure 7 is a graphical representation of address determination; Figure 8 shows further connection point diagrams; Figure 9 shows an implementation example for a video memory device; Figure 10 is another graphical representation of address assignment; Figure 11 is a graphical representation for exposure; Figure 12 is a time diagram; and Figure 13 shows an implementation example of the interpolation step.

Figure 1 shows the basic structure of an electronic light stacking device for stacking entire newspaper pages, briefly full pages, from stored letters and characters by illuminating the drawing medium pixels and line by frame with a light beam 6,77180 moving relative to the drawing medium. Such a light stacking device consists of a control circuit 1 which converts the text to be stacked into a video signal U, and v an exposure unit 2 controlled by a video signal.

From the external text data source 3, in which the text data of the line-separated text is already stored by the typesetting processor, the text information required for full-page exposure is written via data bus 4 to the text page memory device 5 of the control circuit 1. is designated as letter information, also from the external letter font archive 6 along the data bus 7 to the printer character memory 8 of the control circuit 1. The edge line coding will be explained later according to the graphical representation in Fig. 3.

The text information of the full pages is read from the text page memory device 5 and fed along the data bus 9 to a decoder 10, which decodes the text information into write and position commands as well as character information. Letter instructions, such as letter size, thickness, writing angle, italics, and drawing precision, as well as placement commands for individual lines of text, are passed along data bus 11 to video signal generator 12. call as characters, which characters are read for characters from the character memory 8 and passed along the data bus 15 to the video signal generator 12.

The video signal generator 12 generates write commands corresponding to the contour-encoded letter data, taking into account digital switching point signal signals, which are arranged via the data bus 16 to the downlink video memory device 17 according to the pixels and rows.

^ 77180

The exposure unit 2 consists of a light source 19, for example a laser source. The light beam 20 formed in the light source 19 passes through a light modulator 21, which may be, for example, an acousto-optic modulator (AOM), through a hole dimmer 22 and a lens 23 to a reversing mirror 24 and reflected from a polygonal rotating mirror 25 (polygon mirror) perpendicular to its axis 20 against the shaft. The motor 27 rotates the rotary polyhedral mirror 25 by a constant angular velocity in the direction of the arrow. A drawing medium 29 is arranged on the flat-bottomed drawing board 28 in the form of an entire side to be exposed. The drawing medium is, for example, a film material or a photosensitive printing plate. Due to the rotation of the multifaceted rotating mirror 25, the light beam 20 'reflected from the individual mirror faces 30 and focused on the drawing medium 29 by the lens 31 is directed from the picture line in the drawing direction (U-direction) over the drawing medium 29. against (V-direction) by means of motor 32 and screw 33. In this way, the light beam 20 'always scans the drawing medium 29 as complete lines 36 in a drawing grid, the grid width of which depends on the drawing resolution or the distance between the pixels and the lines. One such full page comprises, for example, blocks of text arranged according to a sketch plan, also called articles. The text blocks 34 consist of individual lines of text 35 and lines of text 35 of illuminated characters. Each line of text 35 generally comprises a plurality of lines 36. Since the light beam 20 'always illuminates a full line approximately the entire width of the entire page, each full line 36 consists of sub-lines belonging to different blocks of text 34 or lines of text 35. The video signal generator 12 must therefore be connected. the individual sub-lines into a complete picture line 36 so that the video information of the individual pixels is stored in the memory locations of the video memory device 17 corresponding to the positions of the associated pixels in the overall picture line 36. The exposure of the drawing medium 29 may be positive or negative.

8 77180

For example, in positive exposure, each image line or sub-image line consists of unexposed white dots and exposed black dots. In this case, the characters are formed line by line from the black blocks, whereby during the drawing of the full line 36 only one black block is exposed at a time for each successive character. The lengths of the black and white blocks are determined by the switching time, which is controlled by the video signal U on the conductor 18 by means of the acousto-optic modulator 21, v

The reading of the video signal from the video memory device 17 is synchronized with the relative movement of the light beam 20 'over the drawing medium. For this purpose, a pulse generator 37 is connected to the axis of rotation 26 of the multifaceted rotating mirror 25, which forms a star sequence T1 of pixels. The pixel sequence T1 of the pixels is passed along the conductor 38 to the synchronization stage 39.

Outside the flat drawing support 28 there is a photoelectric pulse generator 40 arranged in the deflection plane of the light beam 20 ', which every time the light beam 20' hits the end of the full line 36 forms a "end of line" synchronization pulse on the conductor 41 and from the synchronization pulse, "end of line" number addresses and read commands to the video memory device 17, which commands are passed to it along the memory and control bus 42. The synchronizing pulses T 1 "end of image line" are applied simultaneously to the forward motion control 43 of the forward motion motor 32 in connection with the stepped forward movement of the flat drawing board 28, which always completes a step to the next image line 36 at the end of the full image line 36. The exposure unit 2 may alternatively consist of a row of LEDs.

In the simplest case, the entire page 29 is exposed in a single pass. The silo is a video memory device 17 formed as a full-page memory arranged in pixels and rows of images, in which video data is stored for all the pixels of the full page 29 in the order required for exposure. Such a full-page memory, the length of a full-page length and the length of a full-width image line, then reproduces the full-page data with pixel resolution.

Preferably, the entire page 29 is exposed in time-sequentially as strip-like blocks, whereby an exposure window having a height h and a width equal to that of the full line 36 is moved over the entire page 29 stepwise up to a height h. This method is shown graphically in more detail in Figure 2a.

Figure 2a shows once again the full page 29 to be exposed with image lines 36 and one line of text 35 with a certain number of lines 36 in the text line. For example, the exposure window 200 has a height h corresponding to q lines 36 and one full line 36 width, p pixels. The position of the exposure window 200 on the entire page is each determined by the vertical coordinate value v „(upper edge of the exposure window). Exposure window 200

F

at position I, the coordinate value of which is vpi, for example, the lines vn-v 1 are exposed, after which the exposure window 200 is moved forward by its height h, regardless of the position of the text line 35, to the position II, the coordinate value of which is vp2, and the corresponding lines are highlighted. Since the displayed text line 35 does not completely coincide with one exposure window 200, this text line is drawn according to the invention with two successive charges, namely the upper part of the text line 35 in position I and the lower part of text line 35 in position II, achieving high drawing speed.

In this case, the video memory device 17 has a sub-memory 88 arranged in pixels and rows, which has a smaller capacity and which is shown only schematically in Fig. 2a. This sub-memory 88 has the same height as the exposure window 200 and has q memory rows vQ-v ^, and has the same width as the full image line 36, so it has a memory capacity p x q for video data. Only the video data belonging to a single location of the exposure window 200 is stored in this sub-memory 88, so that the sub-memory 88 in each case provides only the data of the relevant exposure window 10 771 80 200 on the full page 29. The exposure of the full page 29 must therefore be arranged on the memory lines vQ-v ^ _ ^ the lines at each position of the exposure window 200, for example, at position I of the exposure window 200, the lines v -v η and at line II the lines v -v_, as shown schematically in Figure 2a as 0 q-1 q 2q-1 '.

Preferably, the embodiment is provided with two partial memories 88 and 88 'operating alternately. While reading from the second sub-memory for the current drawing position of the video information and exposure window 200, new video information is already being written to the second sub-memory for the next position of the exposure window 200.

As already mentioned, the video information belonging to the individual locations of the exposure window 200 on the entire page 29 must be stored in and read from the sub-memory in the order required for exposure within that exposure window 200. What text data or video data derived from them is needed at any given time depends on the draft plan of the 29 full-size pages to be drawn.

In the simplest case, the text data stored in the text data source 3 without arranging are arranged in advance in the required order before being exposed to the text page memory device 5, and then converted into video data, which, however, is time consuming. In the implementation example, on the other hand, no pre-sorting is done. Preferably, the text information required therein is searched for the current location of the exposure window 200 during the exposure from a plurality of unsorted text information. In this search operation, the text page memory device 5 next determines which articles and lines of text in each case completely, or only partially, match the current position of the exposure window 200, and then retrieves the correct text information in the order in which they are detected and converted to video information.

This search operation is explained graphically as shown in Figure 2b. Figure 2b shows a sketch plan of the exposed full page 29 with a line of text 35. The search can be imagined as the viewfinder window 202 corresponding to the exposure window 200 (Fig. 2a) on the full page 29 being moved forward over the sketch plan 201, and only the text information of the text lines 35 currently or partially visible in the viewfinder window 202 at the current location of the viewfinder window 202 is retrieved.

To this end, the text data contains article geometry data and line geometry data, which determine the positions of the articles and text lines in the sketch plan 201 with coordinate values u and v. The articles and text lines corresponding to or extending to the respective search window 202 are therefore determined by examining the geometry data. vertical coordinates of the various locations in the search window 202 in the draft plan 201 in the text page memory device.

F

in step 5, for example by comparing the top edge 203 of the text line 35 with the bottom edge 204 (v + h) of the search window 202, and Γ the top edge 205 (v) of the search window 202 with the bottom edge of the text line 35

F

With 207. In the example shown, only a part of the text line 35 hits the search window 202, the vertical dimension Vp of which is obtained as the difference from the coordinate value vQ and the coordinate value vp of the writing base line 207 of the text line 35.

The method and apparatus according to the invention will now be described in more detail.

The first step of the method describes the contours defined by the edge coding of the characters starting from the starting point on each outer side by means of successive rectangles or rectangles and curves, and determines their coordinate values

The edge line coding of the characters and the format of the letter information stored in the character memory 8 of the control circuit 1 will be explained with reference to Fig. 3.

12,771 80

Figure 3 shows a character rectangle 45 with a coding grid 46 corresponding to the highest possible drawing resolution. The encoder raster 46 shows the letter 47, in this example a greatly simplified "O", with a letter base 48. For the edge line encoding of the letter 47, an XY coordinate system 49 is introduced, the X axis and origin of which are at the letter base line 48 of the letter 47. a rectangular character area 50 containing the information 47 itself. The width of the character area 50 corresponds to the width (ZB) of the character and its height is given by y and y, from the letter baseline max min 48. In addition, the leading width (VB) and overall width (GB) are indicated by the character square 45.

In character edge coding, closed border lines (contours) are rotated from a starting point on each contour in a certain direction of rotation and described by successive border pieces whose type, length, and location are given as coded letter information. Border line pieces can be unit vectors, vectors of different lengths, or line pieces and curve pieces. In this case, for example, the outer contour is rotated clockwise and the inner contour counterclockwise.

In a preferred edge line coding, the edge lines, as shown in Figure 3, are described as contour pieces in the form of rectangles and circular pieces. The greatly simplified letter "0" describes both the outer contour from the starting point 51 with the coordinates x and yQ1 and the inner contour from the starting point 52 with the coordinates Xq2 and Yq2 'in the coding directions (rotation in the direction of the arrows) with rectangles 53 with straight pieces 55 and circular pieces 56.

In the letter code, successive border pieces of the same type (straight line, circle piece clockwise, circle piece counterclockwise) are combined in one contour instruction. The contour statement contains the type identifier, information about the number of contour pieces, and the coordinate values f 13 771 80 x and y for the individual contour pieces relative to the XY coordinate system 49, whereby the rectangle and the circle piece

The outline command for a straight line contains the following information: 1. "Straight line" (line); 2. number of straight pieces; 3. the end point coordinates x and y of the first straight line; 4. the end point coordinates x and y of the second straight line, and the end point coordinates x and y of the following straight lines corresponding to the number of straight lines in point 2.

The contour instruction for a circular body, on the other hand, contains the following information: 1. "Circular body clockwise" (CW) or "circular body counterclockwise" (ACW); 2. number of consecutive circular pieces; 3. the coordinates x and y of the starting point of the first circular body as relative coordinates with respect to the center of the circular body; 4. end point coordinates x and y in absolute values: and repeating points 3 and 4 for the following circular pieces corresponding to the number of point 2.

Additional outline commands capture the end of the outline and the end of the entire character.

The outline-encoded letter information of the individual characters stored in the character memory 8 consists of said contour instructions for the individual contour blocks and the identifier information of the character rectangle 45, the following items 1-4 being identifier information and items 5-10 contour instructions.

771 80 14 1. Front width (VB); 2. character width (ZB); 3. overall width (GB); 4. the vertical extent y and y of the character area 50,;

Hiax min 5. the starting point coordinates xQ and yQ of the first contour; 6. first contour contour instructions (straight and circular blocks); 7. contour command "end of contour"; 8. second contour starting point coordinates x and y · 0 0 9. second contour contour instructions (straight and circular blocks); 10. contour command "end of contour"; and repeating steps 5-7 corresponding to the number of outlines, wherein at the end of the last outline of the character, an outline command "end of character" is issued instead of the outline "end of outline" command.

The edge-coded letter information read from the character memory 8 is further processed in the video signal generator 12 / as described in more detail in Figure 4.

The video signal generator 12 comprises a character decoder 57, a conversion stage 58, a calculation stage 59, an interpolation stage 60, an interpretation stage 61, and a switching point memory device 62.

In the video signal generator, in the second step of the method, for each called character, the coordinate values xQ and y ^ of the starting points and the end point coordinates x and y of the contour blocks are changed by a magnitude factor obtained from the plotted character size and the relationship between the plot raster and coding grid. To this end, the edge-coded letter data read from the character memory 8 along the data bus 15 is decoded in the character decoder 57 into length values "front length" (VB), "character length" (ZB), "total length" (GB), and character area 50 length values ymax and on data bus 64, as endpoint coordinate values x and y for individual contour blocks on data bus 65, and as contour instructions for "line" for straight blocks, 15 771 80 "CW" for circular blocks clockwise and "ACW" for circular blocks counterclockwise on the data bus 66.

Letter instructions, drawing accuracy, font size, italics angle, writing angle are passed to the conversion stage 58 along data bus 11. In the conversion stage 58, the length values, the start point coordinates xQ and yQ, and the end point coordinate values x and y of the contour blocks stored in the character memory 8 encoded in the coding raster are changed by said magnitude factor. In addition, the writing angle and italics angle are taken into account when changing the length values and the starting point coordinates xQ and yQ. In the calculation stage 59, the changed character width ZB 'is calculated, as well as the changed values y' and y '. which degree of calculation 59 1 max J 1 min

has brought from the conversion stage 58 along the data bus 67, a modified and possibly still twisted character area 50 '. Calculated starting point coordinates xQ 'and yQ', width ZB 'and the new character area 50' length y '-y' and the modified size-1 max 1 mm J of the character rectangle 45

the female width GB 'is input to the data bus 68. Simultaneously, the difference values bc and 4y between the start and end points of each contour block are calculated from the modified endpoint coordinates x' and y 'of the contour blocks on the data bus 67' in the calculation stage 59, and

Thus, the contour-encoded letter information is made to be converted into video information for the drawing element.

In the third step of the method, the switching points are transmitted to the exposure unit 2 in the form of video data required for drawing the characters. To this end, the pixel points in the drawing grid that are as close as possible to the contour blocks are then determined by stepwise interpolation between the intersection points of the drawing raster for the individual successive contour blocks of the character. In connection with the interpolation, the successive interpolation steps are checked in each case with respect to the change of direction, and the pixels between two successive interpolation steps 16 771 80 are marked as valid switching points by the switching point marking signal in the case of a certain interpolation step combination.

In interpolation, in the interpolation stage 60, each contour block is approached from its starting point to its end point by rectangular interpolation steps + x ^ and + y ^, which are the length of the drawing raster step and parallel to the drawing raster, and / or by diameter steps + X £ / + y ^ ·

The pixel coordinates x * and y * are then obtained from the starting point coordinates x'Q and y 'modified according to Equation (1) and from adding the interpolation steps performed with the correct sign to the pixels.

n X. = x0 '+ IX.

n (1) y * - y0 '+ Jyj

As soon as the endpoint endpoint is reached, an "end-of-paragraph" signal is generated at interpolation stage 60, which is provided along line 70 to the memory control unit 14 of the character memory 8 to obtain the next contour of the character in question for further processing until all characters are read out. wherein the "end of character" signal is transmitted to the character decoder 57 via the conductor 71.

An implementation example of the interpolation stage 60 is shown in Fig. 13. In order to check whether the pixels to be interpolated are valid switching points, the performed interpolation steps + xi and + y ^ are forwarded via conductors 72 and 73 to the interpretation stage 61. and wherein the pixel is marked as a switching point when at least one of the successive interpolation steps occurs vertically against the line direction 17 771 80 and both subsequent interpolation steps occur in the coding direction, i.e. in the case of a certain interpolation combination. Thus, additional connection points are advantageously removed, especially with horizontal contour blocks.

The interpretation stage 61 consists, for example, of two sequentially connected registers 79 and 80 for the intermediate storage of the interpolation steps + x ^ and + y ^ with their associated signs. Registers 79 and 80 synchronize the interpolation stage 60 via conductor 81. In the next section, the interpolation step written to register 79 is taken to register 80, while the next interpolation step is written to register 79 so that registers 79 and 80 always store two consecutive interpolation steps. The outputs 82 and 83 of the registers 79 and 80 are fed to the address inputs 84 and 85 of the hard memory 86, which stores interpolation step combinations for marking pixels as switching points depending on the change in direction of two consecutive interpolation steps in the coding direction. These interpolation step combinations are shown below in the form of a table in which each switching point having a switching point marking value is denoted by the letter "H".

(n + 1) Interpolation path + x. -x .. + y. -yL + xi / + y. + xj / -yi -xi / + yi "xi / _Yi 3 ---—--—-—

M + X. L H L H L H L H

3 x

* -x. H L H L H L H L

U 1

C + Y. HLHLH L H L

-H 1

O-y LHLHL H L H

H 1

g, + x4 / + y. HLHLH L H L

U 1 1

2 + Xi / -Yi LHLHL H L H

• S - ^ / + ^ HLHLH L H L

ö -Xi / -Yi LHLHL H L H

When the fcuv point is detected as a switching point, the solid state memory 86 or the interpretation stage 61 outputs a corresponding switching point signal Iin to the conductor 87.

18 771 80

Figure 5 has sought to clarify the determination of switching points.

The interpolation steps + x ^ and + y ^ along the straight block 74 between the starting point 75 having the coordinates x 'and y' and the end point 76 having the coordinates 0 0 x'0 + Δχ and y'0 + Ay represent those pixels 77 of the drawing screen 78, which approximate as well as possible the straight block 74. In the interpretation stage 61, the pixels 77 to be marked as connection points are shown in Fig. 5 by thicker dots.

This determines all the connection points of the character.

In the fourth step of the method, a switching point image of the character 62 is formed in the switching point memory device 62 by the character parts that each hit the exposure window 200 or the search window 202.

Figure 6 shows such a connection point image (right) for the letter "H" (left).

A ΧΫ coordinate system is arranged to correspond to the memory locations of the switching point memory device 62 so that each memory location can be assigned by the coordinate values x and y. The "width" of the switching point memory device 62 depends on the character size and character rotation, and its "height" is the same as that of the exposure window 200 or the search window 202, i.e., q memory rows y ^ -y ^.

In the simplest case, the switching point memory device 62 is arranged as a bitmap memory, i.e., one pixel or switching point of the character rectangle is arranged for each memory location, which can be indicated by corresponding coordinate values. In a preferred embodiment, a 16-bit data word may be stored at each address to save memory capacity 19,771,180. In this case, the order of the pixels or switching points and addresses is such that each 16-bit data word represents a 4 x 4 pixel or switching point field. The write addresses xgn and ygn required to denote the connection points of the letter or part of the character for the access point memory device 62 in the ΧΫ coordinate system are obtained according to Equation (2) from the pixel coordinates x * and y * in the XY coordinate system obtained from Equation (2) and the part 200 or the current position of the search window 202 in the UV coordinate system.

n * Qr, = xr> '+ Σχ;

Sn ° i i (2) 5Sn - VG - VF - y'o - Pi In this case, the coordinate values x 'and VG “vp“ Y'0 represent the respective starting address for the character or part of the character in the switching point memory device 62.

The geometric similarities are clarified in the graphical representation of Figure 7. Figure 7 shows a sketch plan 201 with a search window 202 in a UV coordinate system with an origin of 208. The search window 202 has a position value Vp. The search window 202 shows a character rectangle 45 and its XY coordinate 49 (Fig. 3) with origin 209. Point 210 represents the starting point with a contour at coordinate values x 'and y' ^ and point 211 an interpolated pixel with pixel coordinates x * and y *, nn where x * = x'0 + and y * e y'Q + £ y ^. The value vQ represents the coordinate value of the letter base line 48 of the character rectangle 45 in the UV coordinate system. Also marked is a ΧΫ coordinate system for the switching point memory device 62 having an origin of 212. The x-axis of the XY coordinate system coincides with the upper edge 205 of the search window 202. When the UV coordinate system of the draft plan 201 or the full page 29 is fixed, the ΧΫ coordinate system moves with the search window 202 or the exposure window 200.

20 7 71 8 0

From the values ygn obtained from Equations (2), it is further checked that y ^> ygn> y, and then accepted as write addresses to the switching point memory device 62 if the calculated values ygn are between the limit values yQ and y ^ ^ which determine the height of the switching point memory device 62 or the height of the search window 202. This check preferably achieves that only the switching points belonging to the part of the character visible in the search window 202 are taken from the fixed access points of the entire character in the access point memory device 62, whereby fast memory filling is achieved.

The structure of the switching point memory device 62 will be described in more detail below.

The switching point memory device 62 has a memory consisting of two separate memory areas, or as in the implementation example, two individual sub-memories 88 and 88 'having address inputs 89 and 89', data inputs 90 and 90 ', data outputs 91 and 91'. The sub-memories 88 and 88 'also operate alternately, i.e., when the switching point marking signals of the switching points fixed in the character interpretation stage 61 are written to the second sub-memory, the switching character marking signals of the previous character written to the second sub-memory are read from this, and the video memory is entered.

A memory control device 93 or 93 'and a limiter 94 or 94' are arranged on each of the sub-memories 88 and 88 '. The switches 94 and 94 'are used to connect the write addresses xgn and ygn to the address line 95 or the read addresses x, depending on the current mode of operation of the partial memory. and yT to the address line 96 via the address lines 97 or 97 'to the corresponding memory control device 93 or 93' and thence along the address line 98 or 98 'to the address input 89 or 89' of the sub-memory 88 or 88 '. The data outputs 91 and 91 'of the sub-memories 88 and 88' are connected by a second interleaver 99 to the data bus 16 depending on the read or write operation of the sub-memories 88 and 88 '. Since the operation of the sub-memories 88 and 88' changes 21 , 94 'and 99 with the signal "end of character" on the other side of the conductor 71, provided that the reading of the second partial memory has already ended. Otherwise, the change will take place at the end of the chapter.

For example, the sub-memories 88 and 88 'are assembled from dynamic RAM units.

The write addresses xgn and ygn for the switch point memory device 62 on the memory line 95 are formed according to Equations (2) in the X memory counter 101 and the Y memory counter 102 formed as forward / backward counters.

To this end, the starting point coordinates x'0 calculated in the conversion stage 58 are provided along the data bus 100 to the reception input 103 of the X pointer 101, and the starting point coordinates y '^ and coordinate values indicating the position of that character with respect to the search window 202 are provided to the data bus 100 Y to receive-input 104.

The synchronization inputs 105 and 106 of the X-address counter 101 and the Y-address counter 102 are connected to the interpolation stage 60 via the conductors 107 and 108. to or subtracted from one of the address counters 101 and 102. The indication created by the X-pointer counter directly forms the writing address x_ of the memory locations to be called in the sub-memories 88 or 88 '. The Y-pointer counter sn 102 includes an additional checking device, by means of which the current indication of the Y-pointer counter is forwarded as a write address only when the indication of the counter is within the limits y described above. and y. between.

Once the interpretation stage 61 has identified a pixel as a switching point, the switching point marking signal is applied from the output of the interpretation stage 61 via a conductor 87 and an interleaver 94 or 94 'to the memory control unit 93 or 93' which is momentarily in the write state and converted to a write command there. 109, the write command stores a logical "H" switching point in the write state of the currently called write address Xgn and 9gn by the control stage 110 connected to the data inputs 90 and 90 'of the sub-memories 88 and 88'. as a signal to the switching point in question. In this way, all the connection points of the character or part of the character are marked in the respective sub-memory 88 or 88 'until said sub-memory 88 or 88' is switched to read operation due to the "end of character" signal from the conductor 71. Both writing and reading take place in the Read / Modify / Write sections. In connection with writing, an EX / OR connection is established in the control stage 110 between the data already stored and the data to be written.

In the fifth step of the method, the switching point images stored in the switching point memory device 62 for one letter or part of the character are transferred line by line to the video memory device 17, and stored there in each memory location corresponding to the position of all characters the connection points of the characters are written. At the same time, the switching dot images written from the switching point memory device 62 are supplemented, which contain only the connection points of the outer and inner contours of the characters (Fig. 6) so that all pixels between the leading and trailing edges of the black characters are then defined by »video information" H " 8, in which a switching point image stored in the switching point memory device 62 is shown on the left and a filled switching point image stored in the video memory device 17 on the right.

Fig. 9 shows an implementation example for the video memory device 17. Since the switching point memory device 62 already described in detail in Fig. 4 is operatively connected to the video memory device 23 7 7 1 80 17, the switching point memory device 62 is shown in Fig. 9 once again for better comprehensibility.

A UV coordinate system is arranged in the memory locations of the video memory device 17 so that each memory location can be assigned by the coordinate values ΰ and v.

In one preferred embodiment, the video memory device 17 is also constructed as a swap memory with two partial memories 111 and 111 '. While the switching point marking signals are written from the switching point memory device 62 to one of the sub-memories 111 and 111 ', the video data is read as a control signal from the second sub-memory and applied to the exposure unit 2.

For example, the sub-memories 111 and 111 'are also assembled from dynamic RAM units. The width of each sub-memory 111 and 111 'corresponds to the full image line 36 of the full page 29, the image line having p pixels, and their height q memory rows vQ - the height of the image window 200 with q image rows. Both sub-memories 111 and 111 'can also store switching point marking signals p x q for a pixel.

In the simplest case, the sub-memories 111 and 111 'also consist of bitmap memories.

In a preferred embodiment, a 16-bit data word containing video data of a pixel adjacent by 16 lines can be stored in each address of the sub-memories 111 and 111 '.

Memory controllers 115 and 115 'and interleavers 116 and 116' are provided for the sub-memories 111 and 111 'having address inputs 112 and 112', data inputs 113 and 113 'and data outputs 114 and 114'. The interleavers 116 and 116 'each connect the write addresses CL and v to the address line 117 or the read addresses u1 and v1, which are formed in the synchronization stage 24 7 7 1 80 39, the address line 42 and the interleavers 116 and 116' according to the current operation of the sub-memories 111 and 111 '. to the corresponding memory control device 115 or 115 '.

The read addresses x and x required for writing switching point images from the switching point memory device 62 to the video memory device 17

Lj y for the switching point memory device 62 on the address line 96, and the write addresses G and v for the video memory device 17 on the memory line 117, and the necessary read and write commands on the lines 118 and 119 are generated in the address control device 120.

The address control device 120 generates read addresses x. and y coupling

L L

for the point memory device 62 such that the memory locations are called a memory row from the memory row and within each memory row a pixel by pixel, whereby only the address range corresponding to the actual character area 50 (Fig. 3) of the character is called. Thus, only those memory locations are actually called in which data is actually stored, which in a cost-effective manner reduces the time required to write the connection point images. To mark the address range, the modified length values of the character area 50 are applied to the memory control device 120 along the data bus 68.

The calculation of the write addresses uc and vc for the video memory device 17 will be explained with reference to the graphical representation of Fig. 10. On the left, an XY coordinate system is provided for the switching point memory device 62, in which a character quadrangle 45d is arranged. On the right is a UV coordinate system for the video memory device 17. The video memory device 17 has already written connection point images for three character rectangles 45a, 45b and 45c on the text line 35, and the connection point image for the fourth character rectangle 45d must now be moved from the connection point 62d.

The initialization address G is then obtained in the U-direction (in the direction of the dash uv):

Ugo = ϋ © + Σ GB + VB (3) where uQ is the initial coordinate of the text line 35 on the full page 29, 25 7 7 1 80 GB ^, GB2, GB ^ and GB ^ are the respective total widths of the character rectangle 35, ZGB is the sum of all already written character rectangles 45a, 45b and 45c, the sum of the total widths, in the example GB + GB + GB .., and VB is the front width of the character rectangle 45d to be written at that moment. The initial write address ΰ in turn gives a current write address

SO

U-direction:

S

5S = «SO + * L

and the running write address vg in the V-direction (perpendicular to the image-line direction) is then: is *? L (5>

The required values are provided to the bogie control device 120 via the data bus 68. The access point marking signals read from the read partial memory 88 or 88 'along the data bus 16 are re-encoded in the encoding stage 121 corresponding to the different organization of the access point memory 62 and the video memory device 17, and passed to a next interleaver 126 connected to the first 126 data inputs 125 and 125 '. The second data inputs 127 and 127 'of the logic stages 126 and 126' are connected via data buses 128 and 128 'to the data outputs 114 and 114' of the sub-memories 111 and 111 ', while the data inputs 129 and 127 of the logic stages 126 and 126' are connected. 129 'communicate with the data inputs 113 and 113' of the sub-memories 111 and 111 '. In the logic stages 126 and 126 ', the data to be written can be advantageously combined with the data already stored in the sub-memories 111 and 111', for example without an OR connection, and applied to the respective sub-memories 111 and 111 '. In the logic stages 126 and 126 ', the already described filling of the memory locations between the two switching points also takes place with the video data "H" (Fig. 8).

The data outputs 114 and 114 'of the sub-memories 111 and 111' are passed via data buses 128 and 128 'to another interleaver 130, the output of which is connected via data bus 131 to a serial / parallel 26 771 80 transformer 132. In the serial / parallel transformer 132, each A 16-bit data word into a serial 16 video signal from 16 consecutive pixels with a single frame.

The reading of video data as a video signal Uv to the exposure unit 2 from the read-only memory 111 or 111 'is controlled by the synchronization stage 39. The pixel synchronization period T1 introduced along line 38 is counted to one U-address counter 133 and end "always reset the U-address counter 133. The readings u and v of both address counters 133 and 134 give the current position of the light beam 20 'on the drawing medium 29. The read addresses u ^ and v ^ required to read video data from the video memory device 17 are obtained from the counter readings u and v: u = u L (6) vL = v - vp where "v" again gives the actual position Γ of the exposure window 200 on the drawing medium 29 (Figure 2a).

The read addresses u and v for the sub-memory 111 or 111 'in the read state are provided along the address line 42 to the video memory device 17.

The pixel synchronization period synchronizes along line 135 to the series / parallel converter 132 so that it provides video signals for individual pixels to be exposed synchronously with drawing. The pixel synchronization period is truncated at a division ratio 136 in a ratio of 16: 1, and the truncated synchronization period forms a read command for sub-memories 111 or 111 ', which is passed along line 137 to video memory device 116. V-address counter 134 is set via counter input 138 to the number of image lines that can be stored q. According to the preset number of picture lines 77180, the V-pointer counter 134 in each case produces a signal "memory change" which is applied along the line 139 to the terminals 116, 116 ', 123 and 130 to switch the partial memories 111 and 111' from the write mode to the read mode or vice versa.

Fig. 11 shows, by way of example, line-by-line drawing of the letter "H" (right) on the drawing medium 29 stored on the video memory device 17, the connection point image of this letter (left).

Fig. 12 is a time diagram for explaining time periods for the write and read modes of the switching point memory device 62 together with the video memory device 17.

In order to achieve better comprehensibility, Fig. 12 initially shows once again, in principle, the alternating partial memories 88 and 88 'of the switching point memory device 62, and the alternating operating memories 111 and 111' of the video memory device 17 as shown in Fig. 9, in which the interleavers are described as mechanical switches.

In the first time period t ^ -t ^, by way of example, four-character switching point marking signals are prepared for the first position of the exposure window 200 with the drawing medium, whereby the first character switching point marking signals are then written to the sub-memory 88. The lengths of the boxes shown correspond to the time , which character is then being processed.

At the end of the writing operation of the first character, the switching point marking signals of the second character are already written to the sub-memory 88 '. During writing, the access point mark signals of the first character are already read from the sub-memory 88 and written to the sub-memory 111 of the video memory device 17. This operation is repeated until the access point marking signals of all four characters are stored in the sub-memory 111 of the video memory device 17.

The number required to replace the sub-memories 88 and 88 'in the switching point memory device 62 corresponds to the number of characters to be processed, and the memory exchange is indicated by a cross in the figure.

In the second time period, for example, five-character switching point marking signals for the second position of the exposure window 200 are presented, and alternately written from the sub-memories 88 and 88 'to the video memory device 17 sub-memory 111', while switching point-marking signals from the exposure window 200 drawn.

For example, the third time period t -t ^ processes the three-character switching point marking signals for the third position of the exposure window 200, and writes to the sub-memory 111 of the video memory device 17, while the contents of the sub-memory 111 'of the video memory device 17 are read and plotted. The memory change for the sub-memories 111 and 111 'of the video memory device 17 always follows from the new location of the exposure window 200.

In this connection, the time control of the write and read operations is arranged so that the time required to write the switching point marking signals from the switching point memory device 62 to the video memory device 17 is less than the exposure speed-dependent reading time of the second sub-memory of the video memory device 17.

Figure 13 shows an implementation example for interpolation stage 60 for linear and circular interpolation.

The interpolation stage 60 consists essentially of an X counter 145, a Y counter 146, an F adder / subtractor 147, an F adder / xy subtractor 148, a comparator 149, two increments 150 and 151, an F register 152, an F register 153, an interleaver xy 29 7 7 1 80 154, memory register 155, and control and logic degree 156.

To explain the theory of interpolations of lines and circles, reference is made to the following literature sources: 1. "An improved algorithm for the generation of non-parametric Curves", IEEE transactions on computers, Vol. C-22,

Well. 12, December 1973, pages 1052-1060; and 2. "High-Speed Algorithm for the Generation of Straight Lines and Circular Arcs", IEEE Transactions on Computers, Vol. c-28,

Well. 10, October 1979, pages 728-736.

The operation of the interpolation stage 60 will be explained using linear interpolation (Uin) along the straight line 74 as an example, as shown in FIG. To this end, a command is next given to the control and logic stage 156 along the conductor 66 to perform linear interpolation (Uin). The coordinate difference values x and y between the start point 75 and the end point 76, calculated in the two-complement representation in the two-complement representation in the calculation stage 59 (Fig. 4), and the corresponding signs (x and y) are loaded via the conductor 69 to the X counter 145 and the y counter 146. The resulting signs (xg; yg) are passed to the control and logic stage 156, which issues corresponding commands (+/- F; +/- F) to the adders / subtractors 147 and 148 to convert them to summation or reduction-claim. Depending on the signs (x; y) of the coordinate difference values δχ and Ay, one of the four quadrants is assigned.

s s

At the same time, the F register 152 and the F register 153 x y are cleared so that the error value F = 0.

'n After this initial setting, the interpolation errors Fxn + 1 - * a Fyn + 1 are calculated. "Depending on the marked quadrants, the error is calculated by adding or subtracting Δχ and Ay vir-hearvoon Fn such that Fxn + 1 = Fn ± Δχ and Fyn + 1 = Fn ± Δγ *

The control is always arranged in such a way that one of the calculations increases the error value F and the other decreases it. For example, in the case of the first quadrants, the second adder / subtraction counter is set to the summing mode and the second adder / subtraction counter is set to the subtraction mode. The error values F .IjalF. | Generated in both adders / subtractors 147 and 148 are loaded through the F register 152 and the F register 153 through the incremental rate J I xn + ll JI yn + ll 1 den 150 and 151, which are not required in the case of linear interpolation. xy and temporarily stored in them. The error values Fxn + ^ and F. the absolute values are compared in each case in the comparator 149. Depending on the result of the comparison, which is passed along the conductor 157 to the control and logic stage 156, it is decided in the control and logic stage 156 whether the interpolation step (+ x ^; + y ^) X- or Y- and by what sign this occurs, in which case the corresponding interpolation step is always performed in the direction of the largest error value. Based on the comparison of the error values, it is further decided in the control and logic stage 156 which of the error values Fxn + ^ »F .. will represent the error value F in the next round. yn + 1 n In this case, the termination takes place in such a way that the always smaller of the error values Fxn + ^ / Fyn + 1 is chosen as the new error value F ^. The selection of the smallest error value takes place by means of an interleaver 154, which is connected according to the selection instruction (Fsel) formed in the control and logic stage 156 and temporarily stored in the memory register 155, respectively to the conductor 158. The interpolation steps performed (+ x ^; To the X-address counter 101 and the Y-address counter 102 (Fig. 4), and is used there, as already described in detail, for address calculation for the switching point memory device 62. Since the memory counters 101 and 102 are not part of the interpolation stage 60, they are shown in the figure only by a dashed line.

To generate the "end of paragraph" instruction at the end point 76 of the straight line 74 (Fig. 5), the coordinate difference values Δχ and Ay of the straight line in question are then loaded through the counter inputs 159 and 160 into the X-length counter 161 and the Y-length counter 162. The interpolation steps performed (+ x ^; + y ^) are either counted or subtracted from the length counters 161 and 162, depending on the sign. The instantaneous indications of the length counters 161 and 162 are monitored in the control stage 163 to see if their indication is "zero". If the indication of both counters is "zero", the end point 76 of the straight section 74 is reached, and the control stage 163 issues a command "end of section" along the conductor 70.

Claims (30)

31 771 80 A method for drawing characters, in particular characters, from character information stored on a drawing medium in the context of electronic page stacking, wherein the character information required for drawing is retrieved from text information describing the stacking instructions and converted to a control signal wherein the control signal determines the current duration of the switching on of the drawing element by means of pictograms, characterized in that before drawing a) each closed contour of the mark is described from the starting point on each contour in a certain direction of rotation around the contour by successive contour pieces, and their coordinate values x; y> is determined by the coding raster of the first XY coordinate system defined for the square of the character, and stored as coded character data, and during drawing b) the coded character data of each character to be drawn according to the text data is retrieved, and the coordinate values of the starting points (x, y) - 0 0 coordinate values of the pieces <x; y) modifying by a coefficient of magnitude obtained from the size of the character to be drawn and the relationship between the drawing grid and the coding raster, c) for each pixel coordinates <x *; y *> as close as possible to the contour object in question, d) successive interpolation steps <xA; y ^> is checked for a change of direction, e) each detected pixel in two successive interpolation steps <xA; 32 771 80 (f) there is a switching point memory device (62) arranged in pixels and rows for pixel switching points for a character or part of a sign, in each case by means of a marking signal to the drawing element when the observed change of direction of the respective interpolation steps <tapaht ^ y ^> has taken place; , in which the memory device has a second ΧΫ coordinate system, and in which each memory line and each pixel can be indicated by coordinate values (x; y), g) memory addresses (ie ,,; yg) for the switching point memory device (62) are calculated as pixels of pixels marked as switching points of a character or character part x *, y *) and the position of the character or part of the character to be drawn on the drawing medium (29), and the switching point marking signals are temporarily stored in the computed write addresses (x; y), h) there is a video memory device (17) arranged on at least one pixel of the drawing grid. hitting the picture line we for video data of the parts, the memory device having a third UV coordinate system, in which each memory row and each pixel can be indicated by coordinate values (u; v), i) the access point marking signals of the character or character portion stored in the switching point memory device (62) are written character by line to the memory locations of the video memory device (17) corresponding to the location of the character or portion of the character to be drawn on the drawing medium; 17) in connection with writing to the memory locations, the memory locations between the memory locations indicated in the line direction are simultaneously supplemented with video data, l) read addresses (u; v) are formed for the video memory device (17) Li Li from the coordinate medium (u; v) of the drawing pixels ) for the fourth UV coordinate system, and m) the video data stored in the read addresses (u; v) for drawing the character or part of the character Li L is read and used as control signals for the drawing element, the drawing element being switched on or off in each case off for the duration of the control signal. 33 771 80 A method according to claim 1, characterized in that a) the drawing medium (29) is successively drawn on line-like blocks, hereinafter referred to as exposure windows (200), the width of the exposure window (200) in the image direction corresponding to the width of the page to be stacked; the number (q) of image lines, and wherein the exposure window (200) is in each case shifted by a plotting medium (29) for the line-like drawing; b) the switching point memory device (62) and the video memory device (17) have a number of memory lines corresponding to the number of image lines (q); and c) the memory rows of the switching point memory device (62) and the video memory device (17) are in each case arranged to correspond to the image lines belonging to the current position of the exposure window (200) on the drawing medium (29). Method according to Claim 1 or 2, characterized in that, for each position of the exposure window (200) on the drawing medium (29), only the switching points of the characters or character parts which are completely or partially in question are written to the switching point memory device (62). exposure window (200). A method according to claim 3, wherein the text information is provided with position information for characters on the drawing medium (29) in the form of position coordinates, characterized in that the characters or parts of characters corresponding to the respective exposure window (200) are determined by comparing the position information coordinates with the current exposure window (200). to the position coordinates in the UV coordinate system. Method according to any one of claims 1 to 4, characterized in that the pixel is marked as a switching point on the contour when at least one of the successive interpolation steps (x ^, y ^) is perpendicular to the direction of the image line and both subsequent interpolation steps (x ^, y ^) ) are in the direction of contour rotation. 34 771 80 Method according to any one of claims 1 to 5, characterized in that when writing the switching point marking signals from the switching point memory device (72) to the video memory device (17), the memory locations of the video memory device (17) are simultaneously filled with video data. are marked as connection points on the front edges of the black image areas intersected by the lines of the characters or character parts, and between the memory locations marked as connection points on the rear edges of the black image areas in question. Method according to any one of claims 1 to 6, characterized in that a) the coordinate difference values (Ax; Ay) between the start point and the end point of the contour body are determined, b) the interpolation steps (x ^; ^ i) performed in connection with the interpolation are added together according to their signs, and c) generating a signal "end of piece" for the contour piece when the sums of the performed interpolation steps (x ^; y ^) are the same as the coordinate difference values (Δχ; Ay). Method according to any one of claims 1 to 7, characterized in that a) the switching point memory device (62) consists of two alternating sub-memories (88; 88 '), b) the switching point marking signals of the character or character part are written from the second sub-memory as video data to the video memory device (17), and c) simultaneously writing the switching point marking signals of the next character or character portion to the second sub-memory. Method according to any one of claims 1 to 8, characterized in that a) the video memory device (17) consists of two alternating partial memories (111; 111 '), b) the video data of the characters or character parts to be drawn for one position of the exposure window (200) is read from the second sub-memory, 35 771 80 c) simultaneously writing the access point marking signals of the characters or signal parts for the next position of the exposure window (200) from the access point memory device (62) as video data to the second sub-memory of the video memory device (17). Method according to any one of claims 1 to 9, characterized in that the write addresses (Xgn; ygn) required in the system for writing the signals of the access point memory (62) of the character or signal part to the access point memory (62) are calculated from the changed coordinate values (x'0; y ' 0), the sum of the n interpolation steps (x ^; y ^) performed at the respective switching points and the position of the mark and the exposure window (200) perpendicular to the direction of the image line according to the following equations: 5Sn * x'o + “i Vsn * Vg - VF - y'o · iYl where ΐ'ν "is the coordinate value of the character baseline (48) of that character and" Vp "is the coordinate value of the upper edge of the exposure window (200) perpendicular to the direction of the image line in the UV coordinate system, and where" x '"and" v - v - y '"represent the current OJ gf O start addresses. A method according to any one of claims 1 to 10, characterized in that the read addresses (x 1; yL) in the XY coordinate system for the switching point memory device (62) for writing switching point marking signals to the video memory device (17) are retrieved pixels and lines. A method according to claim 11, wherein the characters are in each case rewritten from the character area to the character square, characterized in that the range of read addresses (x · y.) Retrieved for one character is limited to the respective character of the character. 36 771 80 Method according to any one of claims 1 to 12, characterized in that the write addresses (ug; vg) required for writing the access point marking signals of the character or signal part to the video memory device (17) in the UV coordinate system are calculated from the access point memory device (62) read addresses (x; y) and from the position coordinates of the character or part of the character on the drawing medium (29) according to the following equations: seuraa- = OL + EGB + VB + X. OU u * s - h where "u" is the beginning of the line of text on the drawing medium (29), "GB" is the total width of the character squares, "ZGB" is the sum of the overall widths of the preceding characters or character parts on the dashed line, and "VB" is the front width of the character in question with respect to the UV coordinate system of the drawing medium (29). Method according to one of Claims 1 to 13, characterized in that the read addresses (ur; v.) Required for reading pixels of the video data from the video memory device (17) in the UV coordinate system are calculated from the pixel coordinates (u; v) of the respective pixels with the drawing medium (29) and the exposure window. (200) of the current position coordinate value (v) in the UV coordinate system according to the following equations: * L = U vL = v - vp A method according to any one of claims 1 to 14, characterized in that the contour pieces are rectangular pieces and circular arc pieces clockwise and the circular arc pieces counterclockwise. A method according to any one of claims 1 to 15, characterized in that the direction of rotation is set clockwise on the outer contour of the mark and counterclockwise on the inner contour. 37 771 80 A method according to any one of claims 1 to 16, characterized in that the encoded character information includes identification information as area parameters for characters and character squares, contour instructions and starting point coordinates (xQ; yQ) for each contour. A method according to claim 17, characterized in that successive contour pieces of the same type (straight pieces; circular arc pieces) are collected in one contour instruction, which contains information about the type of contour pieces and the number of similar contour pieces, and the coordinate of each contour piece. Method according to Claim 18, characterized in that the coordinate values of the end point of the rectangle are always assigned to the rectangle, and the coordinate values of the end point and the coordinate values of the start point with respect to the circle center of the circular arc in question are always assigned to the circular arc. A method according to any one of claims 1 to 19, characterized in that the minimum width of the memory row of the switching point memory device (62) corresponds to the width of the largest untwisted mark 1 and the maximum width of the memory row to the width of the largest mark in the twisted position. Method according to one of Claims 1 to 20, characterized in that the width of the memory line of the video memory device (17) corresponds to the width of the image line on the page to be drawn. A method according to any one of claims 1 to 21, characterized in that each part of the mark which hits the exposure window (200) is also drawn in this position of the exposure window (200). Method according to one of Claims 1 to 22, characterized in that the partial memories (88; 88 ') of the switching point memory device (62) and the partial memories (111, 111') of the video memory device (17) are always emptied while the stored data is being read. 38 7 7 1 80 Apparatus for drawing characters, in particular characters, from character data stored on a drawing medium in connection with electronic page loading, comprising: a) a memory device (3; 5) for text data; b) a character memory (8) for storing coded character data connected to a memory device (3); 5) for retrieving the encoded character information with text information, c) a video signal generator (12) connected to the character memory (8) for converting the retrieved encoded character information into video information, d) a control element (2) input to the control signal, e) a cache connected to the video signal generator 12) and a drawing means (2) for temporarily storing video data as control signals, f) an address control device (39) connected to the cache for generating read addresses to the cache depending on the relative position of the drawing member (2) relative to the drawing medium (29), and g) a movable drawing board for the transfer medium (29), characterized in that the video signal generator (12) has the following parts: h) a device (57); 58; 59) for generating coded character data, which device is connected to the cache (8) for generating character difference values and instructions for generating character difference data and instructions between the coordinate values of the starting points (xQ; yQ), the coordinate values (x, y) of the contour pieces, an interpolation stage (60) connected to the device (57; 58; 59) for determining the pixel coordinates (x *, y *) of the pixels on the drawing grid as close as possible to the outline lines of the characters or character parts by means of corresponding interpolation steps (x ^; y ^), j) an interpretation step (61) associated with the interpolation step (60) for examining successive interpolation steps (x 1; a memory device (62) connected to the interpreting stage (61) for storing the switching point marking signals of the detected switching points of the character or character portion, l) a pointer counter (101); 102) connected to a character generating device (57; 58; 59), an interpolation stage (60) and a switching point memory device (62) for generating write addresses (ie; y „) to the switching point memory device (62), m) a video memory device (17) as a cache for video data connected to the switching point memory (62) for writing single character switching point marking signals from the switching point memory device (62) to the memory locations of the video memory device (17) corresponding to the character locations on the drawing medium, and n) an address control device (120); a device (57; 58; 59), a switching point memory device (62) and a video memory device (17) for generating character addresses (x; y) for the switching point memory device Li Lj _ (62) and writing addresses (ug; vg); to the video memory device (17) when writing access point marking signals. Apparatus according to claim 24, characterized in that the device (57; 58; 59) for generating character data consists of: a) a decoder (57) for interpreting the character data code, b) a conversion stage (58) connected to the decoder (57) for changing the coordinate values by a magnitude factor and c) a calculation stage (59) associated with the conversion stage (58). Apparatus according to claim 24 or 25, characterized in that the drawing element (2) comprises a) a laser beam generator (19), b) a control signal controlled modulator (21) for switching the laser beam on or off, and c) a deflection device (25; 27) above the drawing medium (29). 40 771 80 Apparatus according to Claim 26, characterized in that the deflection device (25; 27) is a polygon mirror. Apparatus according to any one of claims 24 to 27, characterized in that the memory device (3; 5) has a comparison device for text information for comparing the position coordinates in the text information of the characters with the position coordinates of the exposure window (200). Apparatus according to any one of claims 24 to 28, characterized in that the switching point memory device (62) consists of two alternating partial memories (88; 88 '). Apparatus according to any one of claims 24 to 29, characterized in that the video memory device (17) consists of two alternating partial memories (111; 111 '). 77180