EP0978091A1

EP0978091A1 - Method for processing items on printed pages

Info

Publication number: EP0978091A1
Application number: EP97925872A
Authority: EP
Inventors: Wilfried SÖKER; Yngve Neumann; Volkmar Funke
Original assignee: Heidelberger Druckmaschinen AG
Current assignee: Heidelberger Druckmaschinen AG
Priority date: 1996-06-12
Filing date: 1997-05-26
Publication date: 2000-02-09
Also published as: DE19623327A1; WO1997048071A1; JPH11513155A

Abstract

The invention describes a method for processing printed pages using a contone map (item delta list), which characterises the items on the printed sheet (pictures, graphics, text). The printed pages that are to be processed are formed as compressed, overlap-free contone maps (delta lists), which are generated, using little memory space, by interpreting the page description data (PostScript) of the printed pages. For every pixel on the printed page, the item delta list contains an item-specific identification code, enabling the items on the printed pages to be classified. By using this information when processing each specific item on the printed page (picture replacement, trapping), it is determined which pixels on the printed sheet are to be processed. In this way, a printed sheet can be subsequently processed without changing or interpreting again the original PostScript data.

Description

Process for processing objects on printed pages

The invention relates to the field of electronic reproduction technology and relates to a method for processing objects such as images and graphic elements on printed pages which are available as high-resolution contone maps.

In reproduction technology, print templates are created for print pages that contain all elements to be printed, such as texts, graphics and images. 1 shows an example of a printed page. For colored printing, a separate print template is created for each printing color, which contains all elements that are printed in the respective color. For four-color printing, these are the printing inks cyan, magenta, yellow and black (C, M, Y, K). In special cases such as packaging printing, additional printing inks can be added, e.g. Gold, silver, chocolate brown, etc. The printing templates separated according to printing inks are also called color separations. The print templates are usually rasterized

(Screening) and exposed in high resolution on films, which are then further processed to produce the printing forms (printing plates, printing cylinders). Alternatively, the printing templates can also be exposed directly on printing plates in special recorders. In order to check the content and the colors of the printed pages, print templates in proof recorders are exposed with a recording process which simulates the printing process in a colored output.

The workflow previously used predominantly according to the prior art for the exposure of print templates for printed pages which have been generated in the page description language PostScript is shown in FIG. 2. PostScript data (1) are fed to a raster image processor (RIP) (2), which can be a computer specially optimized for this task or a program on a standard computer. Normally, the separated PostScript data (1) for each color separation of a printed page is generated in a pre-process and forwarded to the RIP (2) (separated PostScript). Alternatively, a colored print page can also be generated in a single PostScript database (composite PostScript). The case of the separated PostScript data (1) is explained further below.

In a first step, the PostScript data (1) is analyzed in an interpreter (3) and broken down into a sequence of simple graphic objects. For this purpose, the artwork is divided into horizontal strips (tapes) which are processed one after the other. 3 shows a tape section (9) with some of the interpreters witnessed objects. The tape section (9) is divided into recording pixels (10). In the example of FIG. 3, the band section is 8 pixels high, numbered from 0 to 7, and 32 pixels wide, numbered from 0 to 31. The resolution can be symmetrical (the same in the horizontal and vertical directions) or also asymmetrical, for example ho - horizontally twice as large as vertically. Objects A to E (11, 12, 13, 14, 15) describe subsegments of text, graphic or image elements which fall into the band cutout (9).

Objects A to E (11, 12, 13, 14, 15) are output by the interpreter in a data format which is referred to as the display list (4) (FIG. 2). The data format describes for each object its geometric shape and the gray value with which it is filled. Objects A to E (11, 12, 13, 14, 15) appear one after the other in the display list (4) in the order in which the associated side elements are described in the PostScript data. Objects that appear later in the display list (4) can partially or completely cover objects that previously appeared in the display list (4). In the example of FIG. 3, object A (11) is partially covered by object B (12). Objects D (14) and E (15) also cover object C (13).

In the RIP (2) the display list (4) is fed in a further step to a raster generator (5) which converts the objects of the display list (4) one after the other into areas filled with raster points and as bitmap data (6) writes to a bitmap memory (7). The grid point size is varied depending on the gray value of the object in the display list (4). The bitmap data (6) of objects that appear later in the display list (4) each overwrite the corresponding areas of the bitmap memory (7). After all objects of a tape have been rasterized by the raster generator (5) and written into the bitmap memory (7), the content of the bitmap memory (7) is forwarded as control signal values to the recorder (8) and exposed there.

This conventional way of working has the disadvantage that with complex contents of the printed page, the interpretation of the PostScript data for certain page sections can take so long that the subsequent work steps (screening, exposure) have to wait for the interpretation to be completed. This is particularly the case when many objects overlap in a band section. Then the bitmap data must be generated for all objects, of which later only the top layer of the overlays, ie only a small part, is needed for the exposure. Therefore, a recorder is required for this way of working. loaned, which can stop and start again during the exposure if necessary. Such a recorder places very high demands on the mechanical and optical precision of its construction and is therefore complex and expensive.

In the German patent application by the applicant "Process for generating a contone map", file number 195 13 105.3, and in the associated PCT application, file number PCT / DE 96/00585, an improved workflow for the interpretation and Exposure of PostScript data described, in which a contone map is generated as an intermediate format in a data format called delta list. The contone map contains only one gray value for each exposure pixel and is therefore free of overlays. It is also data-compressed and can be buffered before rasterization and exposure with little space requirement. This has the advantage that the screening and exposure can take place at high speed regardless of the interpretation of the PostScript data, the recorder running continuously without start-stop operation during the exposure of a printed page and therefore can be constructed more simply and inexpensively.

Figure 4 shows this improved workflow. The PostScript data (1) that describe the content of the print template are fed to the RIP (2), where they are analyzed in a first step by the interpreter (3) and converted into a display list (4) , as previously explained. In a second step, the delta list generator (16) generates the overlay-free contone map of the delta list (17) from the display list and e.g. stored on a disk memory (18). If the print pages are to be exposed, the saved delta lists of the print templates, e.g. the different color separations of a printed page, called up at a later point in time from the disk memory (18), converted into bitmap data (6) by the raster generator (5) and exposed in the recorder (8). The rasterization of the delta list is done in keeping with the speed of the recorder.

The previously described work processes have the disadvantage that they do not allow post-processing of the printed pages after interpreting the PostScript data. Such post-processing is desirable in order to make the last changes before printing begins. This should be done quickly and inexpensively. Such desired last changes are, for example, the exchange of an image, a graphic or a text in the printed page in order to correct a mistake which was recognized late or to use a more current image. Another typical post-processing is the exchange of the raster process for a certain image, which produces a disturbing moire with the raster process initially selected. Such moiré phenomena occur when the image contains very fine structures, for example a striped pattern in a blouse when taking pictures of fashion. In such a case, the screening process for this image should be replaced, for example, by frequency-modulated screening, which does not produce any moire with the structures in the image.

Another typical post-processing is trapping, i.e. the creation of protruding edges for some of the color separations at the borders where colored side objects touch. When printing one above the other, the color separations in the printing press can shift somewhat against one another (register errors). This can result in narrow white gaps at the borders between colored side objects, which are very noticeable and annoying. The protruding trapping edges ensure that, despite the shift in the color separations, some of the colors always overlap, so that the white gaps cannot arise. In principle, it would be possible to incorporate the trapping edges into the PostScript data when designing a print page. However, this is not desirable since the rules according to which the trapping margins are created, their width, etc., depend on the properties of the printing press, but which are not always known at the time of the first draft of a printed page. Trapping is therefore a task which must only be carried out immediately before the print templates are issued, if it is known on which printing press the pages are to be printed.

Finally, it may be necessary to assemble several partial pages, which are available as contone maps, to a new printed page in a post-processing stage. be completed advertisements that have arisen from another production route and are therefore not available in the form of PostScript data.

Postprocessing of the type described is not possible according to the prior art with the workflow according to FIG. 2, since from the interpretation of the post-script data to the exposure of the rasterized bitmap data, no intermediate results are used which are used for postprocessing could. Basically, it would be possible to save the bitmap data before exposure, but the amount of data to be saved would be for the typical resolution of recorders for artwork (e.g. 1333 pixels / cm) very large. 1108 MByte would have to be stored for the bitmap data of the four printing colors of a DIN A3 page, so that the storage becomes complex and expensive, especially if several pages of a brochure, a catalog, etc. have to be stored. Post-processing of the printed page therefore requires the corresponding change in the PostScript data and the repeated interpretation, rasterization and exposure of this changed data.

In the improved workflow according to FIG. 4 the contone maps (delta lists) of the print templates are temporarily stored, but the delta lists only contain information for each pixel about its gray value and with which raster method it is used in the bitmap -Data to be implemented. The information about which pixel of the delta list belongs to which original page object (image, graphic, text) is no longer contained therein. This information about the location of the object boundaries is required, e.g. Exchanging an image in the delta list or finding the limits for trapping where colored side objects touch. For trapping, you also need information about whether a color border matches the edge of a page object. No trapping borders may be created for color boundaries within an image. When retrofitting partial sides, it must be known where on one side there is a surface that is not yet occupied (transparent "hole") into which the other partial side is to be inserted appropriately. Since all this information is missing in the stored delta lists, post-processing related to page objects is not possible in this case either.

It is therefore the object of the present invention to avoid the aforementioned disadvantages and to specify a method with which the postprocessing of compressed and overlay-free contone maps (delta lists) of the printed pages is made possible without the PostScript data change the pages and reinterpret them.

This object is achieved by using a contone map generated in addition to the exposable delta list, which contains the information about the type and the position of the original page objects and which is also referred to as the object delta list. The object delta list serves for object-related post-processing (exchange of images, processing of color boundaries between objects, etc.) to identify the pixels in the exposable delta list that are exchanged or changed Need to become. The invention is described below with reference to Figures 1 to 7.

Show it:

1 shows an example of a print page with text, graphic and image elements (prior art),

2 shows the workflow in the exposure of PostScript data (prior art),

3 shows a section of a band with objects that the interpreter creates (prior art),

4 shows the workflow in the exposure of PostScript data with the generation of the delta list (prior art),

5 shows the subdivision of a print template into bands and zones,

6 shows the workflow in the exposure of PostScript data with the generation of the exposable delta lists and the object delta lists as well as the post-processing on the basis of the delta lists and

7 shows an example of the content of a print template and the content of the associated object delta list.

General

In the German patent application by the applicant "Process for generating a contone map", file number 195 13 105.3, and in the associated PCT application, file number PCT / DE 96/00585, the generation of an exposable contone map (delta List) from the PostScript data of a printed page. At this point, this is therefore only explained to the extent necessary for understanding the method according to the invention for generating an object delta list on the basis of a contone map. An exposed contone map describes a print template to be reproduced in the form of gray values in which a gray value is assigned to each pixel. The account map is generated from the page description data (PostScript data) of the print page to be reproduced. The gray values of the contone map can be used directly to control the recorder if the recording process can reproduce continuous tone values, such as a proof output device. For recording processes that can only reproduce two tone values (white or black), the gray values are converted in a raster generator, which is connected upstream of the recorder, into raster points, with which the gray values are simulated for the eye. In the recorder, the printing originals are exposed to the recording material pixel by line and line by line using at least one exposure beam. During the exposure, the control signal values determine which pixels are exposed as parts of the halftone dots or are not exposed by the control signal values switching the exposure beam on and off accordingly.

For the preparation of the delta list, the overlaps of the objects in the display list (FIG. 3) are suitably eliminated and the data is then compressed as much as possible. The delta list is free of overlaps because there is only one gray value in the delta list for each pixel. When choosing the compression method, a compromise must be found between a high compression factor, fast compression and, above all, very fast decompression.

The delta list essentially contains gray values and raster information which can be converted and output into bitmap data by a raster generator, keeping pace with the recorder speed.

The generation of the delta list and the screening can be carried out with different resolutions. An advantageous variant is e.g. the calculation of the delta list with 666.5 pixels / cm and the screening of the gray values with 1333 pixels / cm. The screening can also be asymmetrical, for example with 2666 pixels / cm in the line direction and 1333 pixels / cm perpendicular to the line direction.

The data format of the delta list is byte-oriented. Each byte is a command, which in some cases is followed by data bytes. The coding of the commands is chosen in such a way that the highest possible compression of the data is achieved. On arrival Each delta list contains general information, such as the length of the delta list and the length of a scan line. In addition, the delta list contains information about the raster process (screening), according to which the objects are to be converted into bitmaps by the raster generator.

Since very different page contents with different compression properties can occur in different parts of a print page, the print page is divided into horizontal strips (bands) when generating the delta list, and these are further divided into successive sections (zones). Optimized compression methods can then be used in the bands and zones.

Fig. 5 shows the division of a print template (19) into bands (20) and zones (21). The height of the bands and the width of the zones is arbitrary, but it is advantageous for processing if the bands are all of the same height and the zones are all of the same width. It is also advantageous if the band height and the zone width are powers of 2.

Since often large parts of the information on a print page consist of a few different gray values, e.g. only from black / white information (text), gray values are coded in the delta list with different number of bits, e.g. 1 bit / gray value for black / white information and 8 bits per gray value for contone information. This measure also contributes to the compression of the delta list.

The compression of the data in the data format of the delta list is based on the runlength method, which is modified for the special requirements. Command bytes exist in the data stream, which can be accompanied by a run length and / or one or more gray values. The compression also takes into account repetitions of the entire content of a zone in the Y direction, the X direction being the main scanning direction and the Y direction being the secondary scanning direction. The following table explains some delta list commands and their coding, which are important for understanding the generation of the delta list. Start of a new volume:

LHD_BAND Byte 0

0000 0011

Start a new line in the band:

LHD_START Bvte O 0000 0001

Start a new zone in the line:

LHD_ZONE Byte 0 Byte 1 Byte 2 0000 0010 Y-cmpr bits

Y-cmpr = number of repetitions in the Y direction bits = number of bits per gray value (1, 8, 12)

Selection of a screening method:

LHD_SCREEN Byte 0 Byte 1

0000 1000 screen index

Screenindex = number of the screening process for the following gray values

Short barrel length:

LHD_REPEATS Byte 0 Byte 1

Ol nn nnnn gray value

The gray value is repeated (nnnnnn + 1) times. Long barrel length:

LHD_REPEAT Byte 0 Byte 1 Byte 2

0001 nnnn kkkk kkkk gray value

The gray value is repeated ([nnnn] x256 + [kkkk kkkk] +1) times.

Uncompressed data:

LHD_UCDATA Byte 0 Byte 1 Byte 2

001 n nnnn gray value 1 gray value 2

This is followed by (nnnnn + 1) uncompressed gray values.

The first byte or the first bits in the first byte of each command are on

Indicator for which command it is and how many bytes with parameters follow for the command. This structure ensures that when the delta list is decoded, each command can be uniquely recognized and correctly interpreted.

Every new band is started with the command LHD_BAND and every new line within the band with the command LHD_START. At the beginning of each zone in the line there is the command LHD_ZONE, in which the parameter "Y-cmpr" is used to encode the number of lines over which the content of this zone is repeated in the Y direction. The parameter "bits" specifies the number of bits with which the gray values are encoded within the zone, e.g. 1 bit for black / white information, 8 bits for contone information with normal gradation (256 steps) and 12 bit for contone information with finer gradation (4096 steps).

The LHD_SCREEN command is used to select a screening method that is identified by the "Screenindex" parameter. With the selected screening method, the screen generator should screen all the following gray values in the delta list until a new screening method is selected again. The parameters of the raster methods such as raster width, raster angle, raster dot shape are stored under the number "screen index" in the raster generator, or they are added to the generated delta list with further delta list commands. A run length of repeating gray values within a zone is described with the commands LHD_REPEATS or LHD_REPEAT. In the LHD_REPEATS command, a 6-bit binary number [nnnnnn] encodes a run length between 1 and 64 in the first byte, in the LHD_REPEAT command, a run length between 1 and 4096 is encoded by a 12-bit binary number ([nnnn] in the first byte and [ kkkk kkkk] in the second byte). The last byte of these commands specifies the gray value to be repeated.

If successive gray values in the line are not the same and therefore cannot be compressed with a run length, such a sequence is described with the command LHDJJCDATA. A 5-bit binary number [nnnnn] in the first byte indicates how many uncompressed gray values follow.

When creating the delta list, the lines of a band are processed from top to bottom, and the zones of a line from left to right. The commands and run lengths generated are closely packed together, i.e. for the zones for which no run lengths are created, nothing is entered in the delta list. Based on the code for the compression in the Y direction in the LHD_ZONE command, the raster generator can decode the delta list so that the run lengths are assigned to the correct zones again.

The creation of the object delta list

6 shows the workflow according to the present invention, an object delta list being generated in addition to an exposable delta list for a print template. As already explained in FIG. 4, the PostScript data (1) of the print preview are converted in a RIP (2) by the interpreter (3) into a display list (4), from which the delta list generator ( 16) an exposable delta list (17) is generated. In addition, the delta list generator (16) also generates an object delta list (22) which contains the information about the type (images, graphics, texts) and the position of the page objects which the gray values in the illuminable delta contain List (17) can be assigned. Both delta lists are buffered for further processing, e.g. on a disk storage (18).

For the post-processing tasks (exchange of page objects, generation of trapping edges, etc.) in a post-processing workstation (23) the illuminable delta lists (17) and the associated object delta lists (22) are fed to a suitable postprocessing method (24). Depending on the type of postprocessing, it may be necessary to process more than one exposable delta list and associated object delta list. If an image is to be exchanged or the trapping edges are to be generated for adjacent page objects, the delta lists of all color separations on the printed page are required. If two or more partial pages are to be combined to form a new printed page, the delta lists of all color separations of all partial pages are required.

As a result of the post-processing, modified illuminable delta lists (25) and possibly also modified object delta lists (26) are created, which are cached again, e.g. on a disk storage (27). If necessary, the modified delta lists can be subjected to further post-processing, e.g. a correction of the first post-processing. Separate computer systems need not necessarily be provided for the RIP functions and for postprocessing; they can also be executed on a computer system.

In the further workflow, the modified illuminable delta lists (25) are fed to the raster generator (5), which converts them into rasterized bitmap data (6) according to the raster information contained in the delta lists and sends them to the recorder (8) Exposure passes.

The same data format described above is used for the object delta list as for the exposable delta lists, ie gray values and values for the screen index, also coded in run lengths. In contrast to the exposed delta lists, the gray values and the screen index values in the object delta list have a different meaning. Each page object is assigned a different combination of screen index value and gray value as the object number, and all pixels which are occupied by the page object in the print template receive this combination of screen index value and gray value in the object delta list, ie the associated object number. Since all pixels of an object in the object delta list receive the same gray value, the run length coding results in high data compression and thus only a small memory requirement for the object delta lists. On the basis of the information about the object numbers, it is possible to look up the object to which each pixel belongs in the associated object delta list when postprocessing exposable delta lists. 7a and 7b show this using an example. The print template (28) in Fig. 7a contains the image (29) and the image (30) and a graphic element (31) with a constant color. 7b shows the content of the associated object delta list (32). The area occupied by the image (29) in the print template contains an area (33) of the same size and outline in the object delta list, in which all pixels are filled with the same screen index and gray value, e.g. with the screen index = 128 and the gray value = 0. Likewise, at the position of the image (30) in the object delta list there is an area (34) of the same size, which is filled with the screen index = 128 and the gray value = 1, for example. For the graphic object (31), the object delta list contains an equivalent area (35) which is filled, for example, with the screen index = 1 and the gray value = 0.

The assignment of screen index and gray value to an object is arbitrary. The following table shows a possible assignment as an example.

Object screen index gray value transparent area 0 0

Graphic 1 0

Image mask 2 0

1 bit - picture 64 - 127 0 - 255

8 bit picture 128 - 255 0 - 255

An "image mask" is a PostScript object that is in the form of bitmap data and in which a fixed gray value or color value is assigned to the "1 bits" while the "0 bits" remain empty, i.e. There are "holes" in the image mask.

In the example assignment above, all transparent areas, all graphic objects (= areas covered with a constant color) and all image mask objects each have the same combination of screen index and gray value, ie only the for these objects is shown in the object delta list Object type marked and individual objects are not differentiated. The assignment can of course also be chosen such that each individual object is given its own object number, which is coded as a combination of screen index and gray value. Whether individual objects or only the object type are to be differentiated in the object delta list depends on the degree of distinctness required for the intended postprocessing steps. In the example above, a different combination of screen Index and gray value are provided as the object number, ie all images can be differentiated individually and processed individually using the object delta list.

Claims

claims

1. A method for processing objects on printed pages which are present as digital data in the form of pixel and line-ordered contone maps (delta lists), characterized in that a further contone map for identifying those present on the printed page Objects is created (object delta list) in which all pixels belonging to an object are given an object-specific identifier.

2. The method according to claim 1, characterized in that the object-specific identifier identifies the type of the object.

3. The method according to claim 1 or 2, characterized in that the object-specific identifier identifies each individual object differently.

4. The method according to any one of claims 1 to 3, characterized in that the further contone map (object delta list) is generated by a programmed page description of the content of the printed page, consisting of image, graphic and text information , is processed by an interpreter, the objects on the printed page are identified, the objects are converted into pixels and the object-specific identifier is assigned to the pixels of each object.

5. The method according to any one of claims 1 to 4, characterized in that the objects in the further contone map (object delta list) are free of superimposition, i.e. each pixel is assigned the identifier for exactly one object.

6. The method according to any one of claims 1 to 5, characterized in that in an object-related processing of the print page based on the object-specific identifiers in the further contone map (object delta list) it is determined which pixels of the print page are edit and which are not.

7. The method according to any one of claims 1 to 6, characterized in that the object-specific identifier is a number for a screening process (Screen index value), a gray value or a combination of screen index value and gray value.

8. The method according to any one of claims 1 to 7, characterized in that the further contone map (object delta list) is data-compressed after a runlength coding.

9. The method according to any one of claims 1 to 8, characterized in that the further contone map (object delta list) is data-compressed by reducing the number of bits per gray value.

10. The method according to any one of claims 1 to 9, characterized in that the further contone map (object delta list) is data-compressed by difference coding between the gray values of adjacent pixels.