JPH06501825A - Gathered dot dither method and device for laser printer - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed description of the invention] Collective don't dither method and device technology for laser printers The present invention can be applied to a two-level graphic processing device such as a laser printer to produce a gray scale image. Method and apparatus for generating values or halftone values! Regarding. More details relates to a method and apparatus for performing dither processing of clustered dots in a laser printer, etc. , especially those where it is not possible to print out discrete output pixels individually. A dithering method for laser printers with output pixels of different sizes in the horizontal and vertical directions. Relates to laws and devices.

Conventional technology Simulating continuous tone images in a two-level graphics processor A technique using digital halftone is known. This digital - various techniques related to halftone processing methods called halftoning and spatial dithering. For more information, please refer to "Digital Halftone Processing" published by MIT Press. Halfton-ing) J (R, Ulichney), 198 7).

Spatial dithering methods are classified into two types based on the properties of the dots generated by the dithering technique. It will be done. That is, the distributed dot dither method and the aggregate dot dither method. distributed dots According to the laser method, output pixels, which are discrete, uniformly sized dots, are uniformly distributed in a narrow area. A uniform distribution is used to express gray areas. The set dot dithering method uses Divide the output pixels, which are dots, into small groups or cells, and By collecting the letters, he expresses the gray space.

The classical set-dot dither method is used in analog graphics processing equipment such as offset printing presses. , has been used for about 100 years. However, in digital graphics processing equipment, , if this device can properly reconstruct the disjointed output pixels, then the distributed dots Dithering methods have generally been preferred. Therefore, existing digital dither algorithms Most of them are relatively low-resolution types, such as plasma panels and liquid crystal displays! , CRT Distributed dots in graphic processing devices such as monitors that can precisely control individual output pixels. It's about todiza. This type of graphics processing equipment also requires higher resolution , lasers with precise and expensive optical and mechanical systems capable of restoring separated output pixels. Also includes a printer. Using clustered dot dither on a digital graphics display generally have insufficient physical restoration capabilities to properly display separated output pixels. Two level display devices that cannot be used, e.g. low resolution laser printers or This was the case with thermal printers, etc.

The most common clustered dot dithering method used in laser printers and other This is a dithering method. Regular dithering methods refer to digital halftoning In a way that is always considered when printing, it is the most commonly used The page description language used uses regular dithering techniques for digital halftones. I'm there. The regular dithering method replicates traditional analog aggregate dot dithering techniques. A periodic or regular pattern representing various imitated gray tones. Filling force cells (e.g. classic screens, spiral dots, lines) screen, etc.). Recently, as an alternative to the regular dither method mentioned above, Vuelho ( LVelha) and Miranda Gomez (J,) “Digital halftone processing using spatial embedding curves (Digital H alftoning with 5pace Filling Curves) J (ACM Computer Graphic Processing s), Volume 25, No. 014. July 1991, pp. 81-90) The method uses a non-periodic spatial embedding curve instead of a regular pattern, Filling the individual pixels within the aggregated dot cells.

Regardless of the type of aggregate dot dithering method used, The problem with the combined dot dithering method is the requirement to increase the image resolution, i.e. Requests to increase the number of lines (lpi) and dots per inch (dpi) and reproduction gray Due to the inability to find a solution to the trade-off with the demand for increasing the number of levels, be. In short, the levels between grays generated using a standard regular pattern Increase the size of the cells to increase the number of dots (then the clustered dots themselves become more noticeable). This causes confusion in the low frequency range of the output image. On the other hand, if you increase the overall resolution, The thickness of the lines must be reduced, and the number of gray levels is reduced. the above The trade-off is that smaller, less expensive printers such as desktop printing systems This is a particular problem with printers. Laser printer of the above system, as a typical example 300x300dpi resolution printers such as Canon CX and SX type laser printers. The standard resolution of the linter engine is the effective size of the output cell, in other words, the print The number of gray levels that a computer engine can produce is severely limited.

Another problem with the set dot dither method is that it is difficult to distribute the error to adjacent cells. That's true. Distributed dot design in that the error can be dispersed or diffused. The reason why the dot dither method is preferred to the set dot dither method for digital graphics processing equipment is that The reason is also here. But unfortunately it works fine with distributed dot dithering Error diffusion techniques can also be applied to larger output cells using aggregate dot dithering. It doesn't work very well. Between adjacent pixels with a distance corresponding to the pixel resolution Unlike dispersing the error, it is located at a distance corresponding to the frequency period of the rask scan line. Because the error is distributed between adjacent cells, the resulting image is particularly sensitive to edges and high tones. It tends to smudge or smudge in areas. U.S. Patent No. 4.654.721 discloses an error diffusion method for aggregate dot dithering, according to which the total error are distributed to at least two neighboring cells in arbitrary proportions. The above spatial embedding The curve method stores the total error along the spatial embedding curve within the output cell being processed. The signal is transmitted to the first pixel and diffused. In the above three methods, the total error is Since it is dispersed, there is still a tendency for the resulting image to smear.

The factors behind implementing the clustered dot dithering method are also that the regular dithering method We are basically considering a standard display device with a resolution equal to , i.e. square output pixels. It's in the fact that it's true. However, they can be formed using smaller, cheaper laser printers. Printer control cards, introduced as a way to increase image speed and quality, have non-square Some have a shape pixel resolution. Generally, such printer control cards are A separate dual process is inserted directly into the sonal computer to create a pixel representation of the image. The image quality and image quality of laser printers can be improved by using increasing speed. The above pixel representation image is the printer engine of a laser printer. by directly controlling the modulation of the laser beam. Will be linted out. Laser printer printer by printer control card The engine modulates the laser to match the printer's horizontal resolution to the printer engine's vertical resolution. This will allow you to increase the resolution faster. An example of a printer control card Laser Machinery Co., Ltd., Even Braley, Minnesota, assignee of the present invention. I have an LX6 brochure national printer control card manufactured by Starr. This is The horizontal component of the printer's pixel resolution is 60% depending on the type of laser printer. 0x300dpi can be increased to 1000x400dpi.

Gains can be made by increasing the horizontal resolution of printer control cards that form non-square pixels. The quality of the output image is greatly improved, but the number of output pixels in the page frame buffer is The size is different in the horizontal and vertical directions. As a result, traditional regular dithering In order to use , it is necessary to readjust the input data to the printer's aspect ratio. It's coming. However, even with readjustment, the theoretical pixels of the dither method are still - It is not a pixel with horizontal and vertical size according to the aspect ratio, but a square It can be thought of as a pixel of shape. Theoretical pixels and controls used in the regular disabling method The grayscale image is distorted due to the mismatch with the non-square output pixels obtained on the card. System performance when printing out images is affected. Therefore, a square pixel Printer control using traditional dithering algorithms and techniques devised for Performing digital halftoning on cards is cumbersome and can result in unexpected and undesired results. This will result in an undesirable halftone output image.

A digital figure processing bag that can form dispersed individual output pixels! Then, the distribution dot Dithering method is preferred. This is because the demand for higher image resolution and the gray This is because a solution to the trade-off with the demand for increasing the number of levels can be obtained.

However, the distributed dot dither method cannot generate discrete output pixels individually. It is not suitable for powerful printer engines. Therefore, the discrete output pixels can be printed separately. Other laser printers that cannot be output, especially the output pixel size in the horizontal and vertical directions The advantages of the distributed dot dither method can be obtained even when used with different laser printers. Therefore, a set dot dither method that can do this becomes necessary.

Summary of the invention The present invention is a collection dot device that can generate halftone images on a laser printer or other device. A method and apparatus for analysing. A pattern using a series of regular grayscale patterns. Unlike traditional dithering methods, the present invention uses interactive grayscale patterns to , obtain more gray scale values than the number of output pixels in the aggregate dot cell. present invention has the same number of output pixels that are turned on, but a different number of output pixels when printed out. Output to obtain an interactive grayscale pattern resulting in grayscale values. It takes advantage of the interaction between pixels. The interaction between output pixels is The present invention is particularly useful in laser printers of a type that cannot be formed. types of graphics display devices, especially laser printers with different horizontal and vertical output pixel sizes. It is most effective in linter.

In the present invention, each value is expressed as a gray scale value with a resolution of a certain number of bits. From a continuous tone input image containing multiple digitized input pixels, a digital half A toned output image is obtained. To memorize a pixel representation of a halftone image uses a page frame buffer. This page frame buffer Defined by a number of clustered dot cells, each clustered dot cell has a vane frame buffer. The output pixel group, which is unique to the present invention, is surrounded by a filter. −Constituting a collective dosoto cell For the number of output pixels, the present invention sets a large number of interactive grayscale patterns. It is used for dot dither processing. Each interactive grayscale pattern When applied to a set of dot cells, unique grayscale values can be generated. compilation The computer's processor dithers each input pixel of the continuous-tone input image to Fill with the interactive grayscale pattern closest to the grayscale value of the input pixel. by generating a unique set of dot cells in the page frame buffer. -forms a full-tone output image.

The interactive grayscale pattern of the present invention has a fixed number of output grayscale values. Contains. As many different interactions as output pixels are turned on in the aggregate dot cell. The desired grayscale pattern is determined by the selective location of the output pixels within the dot cell. , producing different output grayscale values. As will be explained in more detail later, the interaction Select output pixels into cluster dot cells to create a grayscale pattern. Selective place! The effect of this is to print out discrete output pixels individually. impossible for laser printers etc., especially the horizontal and vertical size of the output pixels This is most noticeable in the type of equipment manufactured by the company.

In order to make the collective dot dither method of the present invention more effective, a unique error dispersion technique is used. It's being used. When performing aggregate dot dithering, the computer's processor , the gray scale value of the input pixel and the set dot cell is filled for the input pixel. The discrepancy between the grayscale values of the interactive grayscale pattern selected to Determine an error value that represents the fit. This error value is then applied to multiple adjacent input pixels. It is distributed to one pixel arbitrarily selected from among them. In the example, aggregated dot cells are based on Dithering of input pixels to generate them in the frame buffer is performed in a meandering manner. and the above error values are calculated based on the unprocessed and the neighbors located in front and inside of the input pixel being processed. It is distributed to one manually-powered pixel arbitrarily selected from among the four manually-powered pixels.

Using error variance, the resulting halftone image, especially at the edges of the image and in high tones, Smearing and smudging occurs in the area where the A special error variance adjustment method has been added to the error variance process. Part of the error variance process , the computer processor determines the gray scale value of the input pixel being processed and Find the absolute difference between the grayscale value of an arbitrary input pixel and the error value should be distributed. . This absolute difference is used to adjust the distributed error value. As a result, the error variance is most effective in areas of low tonal contrast, where most of the error is distributed and On the other hand, in areas with high tone contrast, especially at the edges of the image, the error is due to the intensity between the two pixels. Since it is adjusted by the degree difference, there is not much dispersion.

If the input data is oversampled (i.e., the sampling density is (larger than the cell density), in this example, the oversampled input data Side cell analysis is optionally used to maintain data integrity. Let's form a partial dot Unlike the traditional set dot dither method, where the set dot dither method into subcells and convert the grayscale value of each subcell into a full-sized set of dots. Using the same interactive grayscale pattern matching technology used for cells Use and analyze.

In an embodiment of the invention, the printer engine of a laser printer is modulated to Using a printer control card to obtain output pixels with different resolutions in direction and vertical direction There is. The aspect ratio of the output pixel in the example is 2.5 (that is, the height is 2.5 times the width). ), and the aggregated dot cells are vertically arranged in a system with a resolution of 400 There are 2 pixels and 10 pixels horizontally. In the page frame buffer, the aggregate dot set The lines are placed on adjacent rask scan lines at equal offsets, forming a diagonal shape, and half The output image of the tone is formed at an angle of 45 degrees in the horizontal direction.

Selectively position output pixels within clustered dot cells to create interactive grayscaling. The effect of the present invention is to print out discrete output pixels individually. In particular, if you have output pixels of different sizes horizontally and vertically, This is most obvious in laser printers and the like. Discrete output of low resolution laser printers Don't think of the inability to print out individual pixels as a problem with dithering. , the present invention improves the physical restoration function of this printer engine to enhance the dithering process. It is used for. The physical restoration capabilities of laser printers are truly binary Nor is it truly gray scale. laser printer printer The engine's physical recovery function determines whether toner will stick to the imaging drum or not. In this sense, it is binary. However, the physical restoration function is limited to the formation of neighboring pixels. Therefore, it can be said that the pixel is formed by superimposing the energy of the spot image. In this sense, in particular, it is possible to perform precise optical and mechanical control over spot images. Not a cheap laser printer, it's not binary.

Due to the superposition of spot images in the printer engine of a laser printer, discrete Traditional dithering methods, which do not allow output pixels to be printed out individually, do not solve this problem. To make a decision, I am adjusting the tone scale. For tone scale adjustment, e.g. This occurs due to the apparent overlap of adjacent pixels. To compensate for the nonlinear strength restoration function, used for

In contrast to this solution, in the present invention, the superposition of adjacent pixels in the spot image is ・Amount of toner attracted to a pixel, used as a means of controlling interaction between pixels This forms a modified output pixel that changes the . Control of such modified output pixel formation is as follows: High speed modulation on printer control card and unequal horizontal and vertical sizes It is particularly suitable for laser printers that produce output pixels with low resolution. square The laser in a printer engine designed for a pixel is faster than the speed of that pixel. Because the toner is modulated faster, the threshold for attracting toner is revised, resulting in higher pixel resolution. It is now possible to create transformed output pixels.

The pixel representation of each horizontal rask scan line produced in the frame buffer is used to control the printer's laser. Laser is turned on in response to on pixel , a generally circular image of laser light is reflected onto the print drum of the printer engine. and the charge is transferred to the drum. The electric charge attracts and picks up the toner on the drum. In the case of an erase white type printer engine, the toner is removed from the drum. . The toner remaining on the drum is transferred to the paper. A square pixel is considered a rectangle. However, in reality, it is normal for it to ooze out from the edges.

In the parent application of the present invention, selectively deforming pixels around vertical dislocation points results in horizontal , non-grayscale anti-aliasing for vertically varying resolution laser printers. It has been disclosed that it is effective for many years. This means that the laser beam on the drum surface In a roughly circular or elliptical area larger than a theoretical rectangular pixel, the distance from the center of focus It is effective because it decreases in proportion to distance. The total force of the laser hitting a given point is Determine the amount of power on the drum at the chain point and use this amount of power to attract the toner to the chain point. It is determined whether the In addition, the spot image created by the laser beam is based on the theoretical larger than a pixel, so the total energy of the laser hitting a point on the drum is This component is obtained by sweeping the laser beam twice or more.

The present invention utilizes the above techniques at high modulation rates of lasers to achieve desired interactive groups. It creates a ray scale pattern. The first one forming the aggregate dot cell For pixels in a horizontal rask scan line, part of the charge overlaps with the next horizontal rask scan line . In this next rask scan line, the laser is applied to the deformed pixel representation, e.g. A laser is turned on to selectively deformed ON pixels. The duration of the laser energy pulse The duration is insufficient to charge the pixel point and attract toner, but the overlap In the area where the to add enough charge to change the boundaries of the area where toner adheres to the drum. . In this way, some of the pixels in the set dot cell are smaller than the height of one pixel. Controlled by range.

The overlapping of toner from the bottom to the top of the raster scan line is the duty cycle of the control signal. In laser printers, the selection of pixels within a set of dot cells is non-linear. control by providing a pseudo duty cycle signal based on the selective deformation The present invention believes that it is desirable to selectively transform the signal. generally nonlinear In laser printers operating at , there are upper and lower limits to the duty cycle of the control signal, and the effects of the present invention apply to these upper and lower limits. actually limited by the limits.

By using the present invention, collective dot dithering can be applied to laser printers, especially for individual dot dithering. Print out discrete output pixels of impossible horizontal and vertical output pixel resolution is realized with different inexpensive, low-resolution laser printers. According to the present invention, minute The advantages of cloth dot dithering are now achieved with a clustered dot approach. Ru. Moreover, according to the collective dot dithering method of the present invention, grayscale patterns can be Traditional offset printing presses and photosetting that require clustered dot output images to produce It becomes possible to form images for character machines.

It is an object of the present invention to eliminate the possibility of printing out discrete output pixels individually. , when used with a laser printer etc. that has different output pixel sizes in the horizontal and vertical directions, The aggregated dot dither method can realize the advantages of the distributed dot dither method in a simple manner. The aim is to provide cost-effective solutions.

It is an object of the present invention to have an output gray scale that is larger than the number of output pixels in the aggregate dot cell. Provides a collective dot dithering method using a grayscale pattern that yields Is Rukoto.

The purpose of the present invention is to provide an error diffusion method that can be applied to the collective dot dither method including the error adjustment method. It is about providing law.

The object of the present invention is to provide a framebar and sofa memo necessary for storing pixel representations of images. You can use a printer control card to increase image quality without increasing the storage capacity of your printer. Provides a set dot dithering method that allows you to print out halftone images It is to be.

The object of the present invention is to provide a low cost, low resolution laser printer with a printer control card. By using it, you can print out high-quality halftone images. The purpose of this invention is to provide a set dot dithering method.

It is also an object of the present invention to provide manual images for traditional offset printing and phototypesetting methods. The object of the present invention is to provide a set dot dither method that can be used to create a set dot dither method.

For the above and other objects of the present invention, please refer to the following drawings, examples, and claims. It will become clear.

Brief description of the drawing Figure 1 shows a printer for generating an image to be printed out by a laser printer. 1 is a block diagram of a laser printing system with a printer control card.

FIG. 2 is a block diagram showing various functions of the printer control card according to the embodiment of the present invention.

Figures 3a and 3b are graphical representations of aggregated dot cells in an embodiment of the present invention. Ru.

Figure 4 shows a set of interactive grayscale patterns in the aggregate dot cell format of Figure 3. An example of this is shown below.

FIG. 5 is a graphical representation of the toner adhesion capabilities of two types of laser printers.

Figures 6a and 6b illustrate the spot images of the two types of laser printers in Figure 5. be.

FIG. 7 is a process flowchart of a dither method according to an embodiment of the present invention.

Figure 8 shows a regular gray pattern of clustered spiral dots according to a conventional example. This is a graphical representation of the

Example Referring first to FIG. A typical personal computer system includes a printer control card for For example, a desktop printing system 10 will be described. desktop The hardware of the printer printing system 10 is connected to a small, inexpensive laser printer 14. It consists of a personal computer 12. This personal computer The computer 12 is an IBM PC, XT/AT, 386, PS/2 manufactured by IBM. Or Apple Manokintonyu II made by Apple Computer, etc. Good too. In addition, the personal computer 12 has a desktop printing program. , for example, Xerox's Ventura bubbler, Aldus's vane membrane. Equipped with OEM desktop publisher from Digital Kaihatsu Co., Ltd. , and graphics processing programs such as Adobe Photon Yop by Adobe Systems. will be provided. The laser printer 14 is available from Hewle Sotobackard's website. Canon CX or SX laser printer, such as the Lasan Esoton Leeds II , Apple Computer's laser writer, or Canon's Canon LB P8-I11, etc., but other types of printer engines can also be used with the present invention. .

During normal operation without using the printer control card, personal computer 1 2 uses the main PC board 16 and serial board 18 to connect the internal components of the laser printer 14. It is connected to the controller 20 via the serial port 18. serial bit string is connected to the laser printer 1 through the R5-232-C serial interface 22. 4 is sent as a command or information representing the image to be drawn. internal control The printer 20 receives the image sent from the personal computer 12 to the laser printer 14. The real focus sequence is decoded to generate a control signal to be sent to laser 24.

When a control signal is given to the laser 24, the printer engine 26 is activated and the internal controller is activated. An image corresponding to the commands and information received by the controller is printed. laser printer For a more detailed explanation of the operation in 14, see Please refer to the User's Manual for the LaserJet II Printer, 1st Printing.

The speed at which instructions and information are transferred through the serial interface and the internal Since there is an inherent limit to the speed of decoding by the controller 20, the laser printer The page throughput of the printer 14 is very low, and the format of the printed information and Depending on the type, it is less than 6 pages per minute. Eliminate this inconvenience and at the same time To enhance the performance of the printer, various printer control cards 30 are installed. It has been put in. For example, Lasermaster in Eden Prairie, Minnesota. This is the company's LX6 Professional. In the embodiment, the printer control The card 30 is located inside the personal computer 12 and is connected via the internal bus 28. , PCClO2 is directly connected. However, this printer control card 30 It may be installed within the laser printer 14 or may be installed inside the laser printer 14. It may be inside the troller 20.

In this embodiment, the printer control card 30 is the desktop printing system 1. Laser printer 14 and personal computer 12 used in The LX6 professional third card has been selected. This LX6 professional The digital control card includes a frame buffer 32, a dual processor 34, a video input It consists of an interface 36. Dual processor 34 is PCClO2 et al. 28, commands and information representative of the image to be printed are received. Bus 28 is parallel Since it is a bus, its transfer rate is much faster than that of the serial port 18. Figure 2 As will be referred to and discussed in more detail below, dual processor 34 is configured to process continuous tone data. halftone to receive digital image information and store it in frame buffer 32; into digital image information. The laser printer 14 prints one page or Once the image-worthy frame is ready to be printed, it is connected via video cable 38. The video ink in the printer controller 30 is connected directly to the laser 24. Through the face 36, the pixel expression of 71/- Mba Sofa 32 is a laser printer] , 4. The internal switch 40 of the printer 14 allows the printer to The controller 30 disables the laser printer's internal controller 20 and II the 24! You will be able to modulate it directly. Thanks to the direct modulation of this laser 24 The printer controller 30 increases the horizontal resolution of the laser printer 14. Can be done.

To learn more about the operation of the LX6 professional controller 30, Please refer to the LX usage name Gorikido by Laser Master Co., Ltd., which is referred to here. stomach.

Next (see Figure 2), how the printer control card 30 according to the embodiment of the present invention Works as a generation system or provides an overview. Print grayscale images When printing out, printer control card 30 receives continuous tone data from computer 12. receive the internal image J(n). The continuous tone digital image J(n) has A cloud of information points representing the image to be printed - each point is a gray scale represented with N-bit resolution. - has a scale value. Here, N represents the number of bits greater than 1. , usually 8 bits. In many cases, a continuous tone digital image J ( n) indicates the information points obtained from the scanned physical image. Alternatively, J(n) is It shows information points generated by a computer graphics processing program. continuous tone Several standard shapes are used to represent the information points that make up the digital image J(n). There are file formats, such as TIFSPICTSEPS and GIF. fruit In the embodiment, input information in a graphic file format is input to the printer control card 30. Before transferring, it is converted to BostScript (trade name) format.

Continuous tone digital image J(n) is processed by printer control card 30, A halftone digital image J(n') is generated. As mentioned above, the hardware The halftone digital image J (n') is then converted into the desired halftone image 1 (x) is sent to the laser printer 14 to obtain (x). The physical restore function A binary number display device, here indicating the system type of the laser printer 14, is used as an input. receives a binary discrete space image represented by a halftone digital image I(n'). A continuous spatial image expressed as a halftone image I(x) is formed. thing The physical reconstruction function varies widely depending on the binary display device, and the interactive gray scale of the present invention The scale value is adjusted to accommodate differences in the physical restoration capabilities of this binary display device. It can be changed.

Continuous tone digital image J (n) is converted to halftone digital image I (n o) The various functional elements of the printer control card 30 that convert adjusting means 40, tone scale adjusting means 42, coloring means 44, halftone There is an input means 46. Retrospective resampling means 40 extracts continuous tone digital images. Matching the resolution of J(n) to the resolution of halftone digital image I(n') It is a digital counter that allows you to Resolution of laser printer 14 used in the present invention is low, so the retrospective resampling means 30 extracts the continuous tone digital image J. Lower the resolution of (n).

If the input data is oversampled (that is, the sampling density subcell density), the present invention optionally performs subcell analysis. , trying to preserve the integrity of oversampled input data, original state. .

In an embodiment, the retrospective resampling means 30 extracts from the predetermined sample by interpolation. Restore the original image sequence and then resample this restored image at the resolution of the aggregate dot cell. a new image with the same resolution as the halftone digital image I(n’). A continuous tone digital image J(n') is generated. Example halftone color The resolution of the digital image I(no) is a dot of 2 pixels vertically and 10 pixels horizontally, or a cell sensor. The size is 1411pi, or the dot size is 3 pixels vertically and 16 pixels horizontally, or the cell size is 90 pixels. 1pi, and varies depending on the resolution required for the image.

Better results can be obtained after the continuous tone digital image J(n’) is properly adjusted. Further, using the tone scale adjustment means 42 and the sharpening means 44 to obtain Adjustments may be made to the data. Typically, the tone scale adjustment means 42 is A non-linear transformation is applied to the continuous tone digital image J (n’), and the half-tone image to be used is In conjunction with the tone-setting method, the tone-scheduling method is unique to the physical restoration function of the laser printer 14. Compensate for changes in control. In most conventional systems, the laser printer's printer To compensate for spot image superimposition in the linter engine, pure black is mapped to pure white. By removing the bright and dark areas of the continuous tone digital image J(n') to be there was. Therefore, traditional solutions to the problem of overlapping dots in printer engines The solution was to reduce the number of grayscale levels produced by the system.

The contrasting means 44 also eliminates fine haze in images that tends to occur during the halftoning process. To pre-compensate, process the data closest to the black or white level. but In the conventional example, this sharpening means 44 is used in the preparation process for the halftoning process. Instead, it was used as part of the process.

In FIGS. 3a and 3b, the clustered dot cells 5 formed by the halftoning means 46 An example of 0 will be described. As shown in FIG. 3a, each cluster dot cell 50 2. in the frame buffer 32. 20 output pixels in a rectangular grid of XIO 5 It is composed of 2. The output pixel 52 has a 2.5 aspect ratio (400X1000 dpi resolution), the aspect ratio of each set of dot cells 50 is also 2. It is 5. When the halftone grating is oriented at a 45° angle to the horizontal, the set The dot cells 50 are shifted by half a cell between each rask scan line. Implementation of the invention The height of the halftone image is easily maintained across the boundaries of the aggregated dot cells. In order to improve the frequency part, in the embodiment, the boundaries of the three aggregated dot cells 50 are A theoretical cluster dot 54 is defined as shown. As a result, the actual set of dots The grayscale values of cells 50 are distributed at the boundaries of each cell. Theoretical set de In order to obtain the Fill in each cluster dot cell 50, and add a semicircle at the bottom of the cluster dot cell 50 and each corner at the top. Form two quarter circles.

See FIG. 4 for an interactive grayscale pattern 60 according to an embodiment of the invention. I will explain it while referring to it. The half-turning means 46 generates a set of grayscale patterns. The gray scale level is larger than the number of pixels in the aggregate dot cell 50 using the dot cell 50. make a bell The output pixels 52 indicated by diagonal lines in FIG. blank output pixels 52 are off pixels, and each grayscale pattern 6 Represents a grayscale value of 0. The dark line 62 indicates which image in the set dot cell 50. Which of the theoretical set dots 54 as illustrated in FIG. 3b does the element relate to? It shows. For comparison, on each gray scale pattern 60, 0 to 25 Grayscale of a continuous tone digital input with 8-bit resolution expressed as a value of 5. indicates the range of values. For example, a gray scale pattern in the range [69-76] is [77-84] The same number of grayscale patterns in the range (5 out of 20) It has an on output pixel 52. The difference between the above two patterns is shown by 64 and 66. arises from the positions of two of the on-output pixels, as shown in FIG. [ 69 to 76], the pixel shown at 64 is one pixel in between. The grayscale pattern ranges from 77 to 84 pixels. , the pixels shown at 66 are adjacent. If pixels 66 are adjacent, the spot The black and white images overlap, and each pixel is fully charged and attracts toner. Pixel 64 is separated If it is, none of the pixels will be charged enough to attract toner, and the aggregated dot cell 50 will not be charged enough to attract toner. Even if the number of pixels turned on is the same, the total amount of toner adhering to the pixel 66 is less than when

As can be seen from the interactive grayscale pattern 60 of FIG. The ray scale pattern values are divided into 32 ranges in total. Figure 8 shows a comparative example. , showing a typical conventional spiral dot gray pattern. Example 2 A square set dot cell of 2 x 5 pixels that covers the same surface area as a x 10 pixel cell. This is the pattern. The present invention generates approximately three times as many gray levels as the conventional example shown in Figure 8. can do. A cell consisting of 2×10 rectangular pixels is defined as a set dot cell 50. Even when used as a conventional spiral dot, the present invention still The theoretically highest 21 gray scale values obtained by the method (4 x 5 pixels and all 1) can provide 50% more gray levels than 1).

Grayscale pattern 60 may be varied in many ways without departing from the invention. I can do it. For example, the size and shape of the aggregated dot cells 50 can be adjusted according to the desired output resolution, It can be changed depending on the number of grayscale values. I will explain in more detail later , the manner in which adjacent output pixels 52 interact also depends on the laser printer 14 printer. It varies depending on the physical and optical properties of Rintaennon. Also, gathering Dotsutose The cell embedding technique for dithering of 50 is also according to the theoretical set of dots 54. Methods other than filling are also available.

Another example of implementing grayscale pattern 60 is to implement grayscale pattern 60. Manpin of the grayscale pattern for the grayscale value expressed by the turn This means changing the method and scope of the search. For example, in an embodiment, a grayscale pattern 68, 4 out of 20 output pixels are turned off, but [252 to 253 pixels] Set equal to the grayscale value of the range. This is the SX printer of the example. Due to the interaction between the pixels of the processor engine, less than 4 pixels are turned off in the above pattern. It is known that the output result is a completely black set of dot cells 50. It is from. Therefore, the gray scale pattern 60 is in the middle of the gray scale range. It is now possible to take advantage of more sophisticated differences in the interactions between each pixel.

Method of mapping grayscale values to grayscale pattern 60 can produce continuous tones with grayscale values of various resolutions by varying the range. Performs both linear and non-linear mapping functions on digital/tal input images It is possible.

Laser printer 14 is used to create interactive grayscale pattern 60. To better understand the physical restoration capabilities of your printer engine and use this knowledge to It is necessary to apply this method to realize a better disabling method for engines. Figure 5, Figure 6a, Figure 6b, the same number of output pixels as printed out by the binary image printer. The present invention is useful for generating multiple grayscale values for aggregate dot cells. Let me explain the background behind this idea.

I/- The light spot image can be printed with different sizes, shapes and power densities. Depending on the type of engine, the drum is loaded with toner in response to specific video control signals. The way it is attached changes. Figure 5 shows a control system with variable duty cycle. This figure shows the responsiveness of two types of laser printers to control signals. Responsiveness is , vertical height of one pixel (vertical axis) % duty cycle of control signal (horizontal axis) It is expressed by the amount of toner accumulated from the bottom to the top of one rask line. Pudding Due to the responsiveness to the printer engine, the toner The accumulation varies generally linearly as the duty cycle of the control signal increases. I can see that it is changing. On the other hand, other printer engines, printer engine B, etc. As shown by its responsiveness, toner accumulation from the bottom to the top of the rask line is , responds non-linearly to the duty cycle of the control signal.

Differences in toner storage capacity with respect to the duty cycle of printer engine control signals Another reason may be that the types of spot images formed by each printer engine are different. There is. As shown in Figure 68, if the energy distribution is a smooth Gaussian distribution, Toner storage capacity is linear and accumulates the "area error" described in the parent application of this application. is the best way. On the other hand, if the energy distribution is as shown in Figure 6b (By applying a laser beam using a more expensive optical system, the central band becomes narrower. (more exponential distribution), the duty cycle of the control signal On the other hand, the toner storage capacity is non-linear. In this case, as area errors accumulate However, since the degree of overlap between vertically adjacent pixels is small, the desired smoothness cannot be achieved. I can't get it.

Returning to FIG. 5, the present invention also provides a non-linear response to the duty cycle of the control signal. A binary image printer with a toner storage capacity of It is also possible to create a large number of grayscale values. As is clear from the drawing, There are upper and lower limits to the duty cycle of the signal, and above and below, the duty cycle of the control signal The toner build-up height has no practical effect. Below the lower limit, the pixel Even if the image is deformed selectively, deformation toward the upper edge of the image cannot be obtained; In this region, no deformation toward the bottom of the image edge is observed. However, the inventor If the control signal has a duty cycle within the upper and lower limits, it will be non-responsive to the control signal. Even for printer engines that exhibit linear toner accumulation capabilities, it is difficult to control toner accumulation. It has been found to be effective in

Referring now to FIG. 7, a preferred embodiment of halftoning means 46 will be described. . The deformed input image J’ (n’) has the resolution, tone scale, and After the appropriate adjustments have been performed, it is supplied to the halftoning means 46. Implementation of the present invention The dithering means 46 according to the example includes a horizontal rask line 5 stored in the frame buffer 32. This is implemented by a dual processor 34 that generates 0. Horizontal rask line from top to bottom Then, the output pixels 52 of each horizontal rask line are processed in the frame buffer 32, and the output pixels 52 of each horizontal rask line are Generally, it is processed in a meandering manner from the beginning to the right, and then from right to left. As long as the image processing method is consistent, the rask line and the output pixel 52 can be moved in any direction. You can process it with Minimizes the calculation effort of dither processing by dual processor 34 In order to Provided. The first table contains grayscale input values with 8-bit resolution. The grayscale pattern assigned to each of the 256 values, represented by 60 is included. The second table shows the assigned grayscale Contains the error value between the pattern and the actual grayscale value corresponding to the input value. ing.

For each deformed input pixel of image J'(n'), in step 70, the dual pro- gram The processor 34 determines the error distribution pixel J' (p'). This is currently being processed It is located below and inside from pixel J' (p). Any direction number from 0 to 3 One of the four adjacent unprocessed pixels is selected according to the number. error distribution pixel By arbitrarily determining the direction in which the error is dispersed in the shape, the concavity that affects the error dispersion can be Convexity and flattening tendencies can be easily avoided. In step 72, pixel J The difference between the error distribution pixel J’ (p) and the error distribution pixel J’ (p’), the grayscale region ( In other words, calculate the value divided by 256). This value is the error distribution pixel J' (p' ) is used to calculate the amount of error to be distributed. In step 74, pixel J ' Determine the dispersion error of (p) by looking at the second table. In step 76, Calculate the dispersion error from the result of step 72 and set it to the error dispersion pixel J” (p’). Add (or subtract). Finally, in step 78, corresponding to J'(p) . The interaction grayscale pattern obtained from the first table is stored in the frame buffer. 32, the corresponding distribution dot cells 50 are filled in.

In another embodiment, not shown in FIG. 7, a continuous tone digital image is , the output resolution determined by the aggregate dot cell 50 (that is, 1411pi is 901pi) If a resolution higher than pi) is represented, dithering by dual processors is The process may be further divided. In this case, each set of dot cells 50 is The analysis for adapting to the scale pattern 60 is performed on the subcells of the aggregated dot cells 50. For example, the analysis can be divided into four subcells of 1×5 pixels each. after that, Each subcell is assigned a gray scale value corresponding to the gray scale value of the continuous tone digital image. to match the rule pattern part. When the input resolution is higher than the output resolution, the cell By the method of the present invention, which divides the dot into sub-cells, the unique An improved method can be obtained.

Although the embodiments have been described above, the present invention can be modified in various ways. The invention resides in the claims rather than in the embodiments described above.

%Bratty Cycle 100% Fig, 7 Conventional example international search report Continuation of front page (72) Inventor Lukis, Lawrence J. United States of America 55372 One Hundred Rylotys Street, Prior Lake, Minnesota #6366 (72) Inventor: Steidel, Leonard Earle United States of America 55372 Rylow Beach Trail Varnish Daburi, Prior Lake, Minnesota 347 am

Claims (20)

[Claims] 1. Gathered dot dithering method for outputting halftone images with a laser printer , the method is characterized in that it consists of the following steps, and the generated The resolution of the halftone output image has an opposite relationship to the number of grayscale patterns G. There is no way (a) Determined as a grayscale value with a resolution of N bits, which is an integer greater than or equal to 2. providing a continuous tone input image having a plurality of digitized input pixels; , (b) A set of dots representing M output pixels greater than 2 and less than N squared. (c) defining a cell in a page frame buffer; (c) defining a cell in a set dot cell; Interactive graphics that can generate unique grayscale values when applied determining a number G greater than the number G of ray scale patterns, M; (d) Dithering each input pixel to produce a single grayscale value representing the grayscale value of that pixel. Generate a set of dot cells filled with an image scale pattern in a page frame buffer. To generate a halftone output image of a continuous tone input image by using processing means for the purpose of 2. The aggregated dot dithering method according to claim 1, wherein the interactive gray scale parameter Turn is a laser printer that cannot print out discrete output pixels individually. , by using the interaction between output pixels, the output is greater than the number of output pixels in the set dot cell. It is characterized by generating grayscale values. 3. In the aggregated dot dithering method according to claim 2, some of the output grayscale values or the number of interactive graces equal to the number of output pixels turned on in the aggregate dot cell. generated by the Kale pattern, the above interactive grayscale pattern is a set Different output gray scales can be achieved by selectively positioning the output pixels within the dot cells. It is characterized by generating a rule value. 4. In the aggregated dot dither method according to claim 1, the output pixels are horizontal and vertical. Characterized by unequal size. 5. In the aggregated dot dither method according to claim 4, each aggregated dot cell has two vertical strokes. It has 10 output pixels horizontally and is positioned to draw a diagonal shape. It is characterized by forming an output image. 6. In the aggregated dot dithering method according to claim 1, the method further comprises the following steps: It is characterized by consisting of: (e) Using a processing means, the gray scale value of the input pixel and the input pixel are determined. , the interactive grayscale pattern chosen to fill the aggregated dot cells. Determine the error value representing the mismatch with the grayscale value of multiple adjacent input pixels. A step of distributing the above error value to one pixel arbitrarily selected from among the pixels. 7. The aggregated dot dither method according to claim 6 further comprises the following steps. A method characterized by: (f) Using a processing means, the gray scale value of the input pixel being processed and the error value are separated. Determine the absolute difference between the grayscale value of an arbitrarily selected input pixel and Step of calculating the error value to be distributed from the absolute difference. 8. In the aggregated dot dithering method according to claim 1, said step (d) comprises: It is characterized by comprising the steps: (d1) Using a processing means to Generates a deformed input image that has the same resolution as the set dot cell from the perspective of the prime numbers. , resampling the continuous tone input image; (d2) adjusting the tone scale of the deformed input image using a processing means; , (d3) sharpening the deformed input image using a processing means, (d4) A processing means is used to perform dither processing on the deformed input image to obtain a set related to the input pixels. The interactive input that most closely approximates the input pixel of the deformed input image to fill the dot cells. Generate a halftone output image by selecting a grayscale pattern Steps to do. 9. 2. The method of claim 1, wherein the resolution of the continuous tone input image is halftone. is higher than the resolution of the output image, and step (d4) is further composed of the following steps. A method characterized in that: (d41) Step of dividing the set dot cell into a plurality of subcells, (d42) Subcell cell to the input pixel of the deformed input image, which has a higher resolution than the halftone output image. step of assigning, (d43) Using the processing means, perform dither processing on the transformed input image, and In order to fill each subcell of the set dot cell related to the input pixel of Select the interactive grayscale pattern that most closely approximates the applied input pixel Generating a halftone output image by: 10. Collective dot dithering for outputting halftone images on laser printers The method is characterized in that it consists of the following steps: (a) Continuous tone with multiple input pixels digitized as grayscale values (b) providing a set of dots each representing a plurality of output pixels; a step of defining multiple cells in the page frame buffer; (c) When applied to clustered dot cells, each generates a unique grayscale value. (d) using processing means to define a grayscale pattern that can Dither processing is applied to each input pixel, and one mutual value representing the gray scale value of the input pixel is Page pages feature unique clustered dot cells filled with active grayscale patterns. halftone output of a continuous tone input image by generating it in a frame buffer a step of generating an image; (e) Use a processing means to separate the grayscale value of the input pixel being processed and the error value. determining the absolute difference between the grayscale value of an arbitrarily selected input pixel to be scattered; A step of calculating an error value to be distributed using the absolute difference. 11. The aggregated dot dithering method according to claim 9, further comprising the following steps. How to arm the features: (f) Use a processing means to separate the grayscale r value of the input pixel being processed and the error value. Determine the absolute difference between the grayscale value of an arbitrarily selected input pixel and Step of calculating the error value to be distributed from the absolute difference. 12. In the aggregated dot dither method according to claim 10, the input pixels are subjected to dither processing. to generate aggregated dot cells in a serpentine raster format in the page frame buffer, The error value can be arbitrarily selected from among the four input pixels before and after the input pixel being processed. It is characterized by being distributed over one selected pixel. Defined as a grayscale value with a resolution of N bits, an integer greater than or equal to 13.2. From a continuous tone input image with multiple digitized input pixels, In a set dot dither device for outputting a halftone image on a printer, The device is characterized in that it consists of the following means, the halftone produced The resolution of the output image is not directly opposite to the number G of gray scale values. Characterized by: Page frame buffer means for pixel display and storage of halftone images; a plurality of collections, each representing M output pixels greater than 2 and less than N squared; Means defined as a set dot cell, when applied to a set dot cell, a unique group The number of interactive grayscale patterns that can generate a rayscale value, with M A means for storing a large number G, which performs dither processing on each input pixel, and filled with an interactive grayscale pattern representing grayscale values; Continuous truncation by generating unique clustered dot cells in the page frame buffer. processing means for creating a halftone output image of a tone input image; 14. In the aggregate dot dither device described in claim 13, the interactive group Rayscale patterns are lasers that cannot print out discrete output pixels individually. The printer uses interactions between output pixels to It is characterized by generating a large number of output grayscale values. 15. 15. The aggregated dot dithering device of claim 14, wherein the output grayscale value Some of the interactive obtained by a grayscale pattern, the interactive grayscale pattern is , due to differences in the selective positioning of the output pixels within the aggregate dot cell. It is characterized by generating an output grayscale value based on the 16. 14. The aggregated dot dither device according to claim 13, wherein the horizontal direction of the output pixel and They are characterized by unequal vertical sizes. 17. 17. The aggregated dot dithering device according to claim 16, wherein each aggregated dot cell is arranged vertically. There are 2 pixels on the side and 10 output pixels on the side, and they are arranged in a diagonal shape. This method is characterized by obtaining an output image of the screen. 18. The aggregated dot dithering device according to claim 13, further comprising: The gray scale value of the input pixel and the set Grayscale of the interactive grayscale pattern chosen to fill the dot cells. Determine an error value that represents the mismatch with the Means for distributing to one cell arbitrarily selected from the set cells. 19. 19. The aggregated dot dithering device of claim 18, further comprising: characterized in that it consists of means: the gray scale value of the input pixel being processed; and the grayscale value of an arbitrary input pixel over which the error value is distributed. means for determining and using said absolute value to calculate a distributed error value. 20. The aggregated dot dithering device according to claim 13, further comprising: characterized by consisting of means: in terms of the number of input pixels, the set dot cell The continuous tone input image is replayed such that a deformed input image with the same resolution as is generated. means for sampling, means for adjusting the tone scale of the transformed input image, A means to sharpen shape input images, If the resolution of the continuous tone input image is less than or equal to the resolution of the halftone output image, In order to fill the set dot cell regarding the input pixel of the shape input image, the nearest The above deformed input by selecting a similar interactive grayscale pattern Gathered dots, a means of dithering an image to form a halftone output image dividing the cell into a plurality of subcells and assigning the subcells to input pixels of the deformed input image; When the resolution of the continuous tone input image is higher than the resolution of the halftone output image , assigned to the subcell to fill each subcell of the set dot cell associated with the pixel. selecting the interactive grayscale pattern that most closely approximates the pixel performs dither processing on the deformed input image to form a halftone output image. means.