Incremental printing of symbolic information – Ink jet – Ejector mechanism
Reexamination Certificate
1999-08-26
2001-05-01
Nguyen, Thinh (Department: 2861)
Incremental printing of symbolic information
Ink jet
Ejector mechanism
Reexamination Certificate
active
06224189
ABSTRACT:
BACKGROUND OF THE INVENTION
A method of generating high quality output on a mixed-resolution printing device using image processing to drive the output at the highest resolution possible. The invention is applicable to the recent generation of xerographic and inkjet marking devices that use mixed-resolution for the various colors. Specifically, the invention is applicable to inkjet printers with integrated multi-resolution printheads and/or multi-path or tandem xerography engines that operate at more than one processing-speed simultaneously.
One particular example for which the method was developed and demonstrated to produce superior results is the case of a color thermal inkjet printer using an integrated printhead with a nominal resolution of 600-dpi for the colors (Cyan, Magenta, and Yellow) and 300-dpi for Black (K). The reason for the mixed-resolution design is to enable high-quality color output (i.e., “photo quality”), yet to reuse the proven 300-dpi black ink technology from the previous generation products.
In said inkjet case, the printhead embodies an integrated design with a number of thermal ejectors, all arranged in a single physical column. In order to facilitate faster printing of monochrome documents, a larger amount (up to about half) of the available ejectors are allocated to the black ink. The black ejectors are etched on a 300-dpi spacing. The remaining ejectors are equally allocated to Cyan, Magenta, and Yellow, and are etched on a 600-dpi spacing.
Since the printhead is fabricated from a single piece of material, the final printing resolution is determined by the timing of the jet firing and the relative movement between the printhead and media. It is obvious that in order to achieve true 600-dpi printing the printhead must be stopped on the 600-dpi grid locations of the media. However, if the black ejectors were allowed to fire simultaneously with the color ejectors on the same grid, then the volume of black ink would be four times larger than that of any single color, and the application of so much ink would flood the page. In addition, the black ejectors would be fired at a rate of four times faster than nominal (300 dpi) operating frequency and this may cause the printhead to overheat and shut down.
A practical method of eliminating the above difficulty, that is—controlling the amount of ink and at the same time preventing the printhead from overheating is to restrict the firing of the black jets to the 300-dpi grid. Typically, this is accomplished via a pre-programmed firing pattern (also known as the mask) that is stored in the printer's non-volatile memory. Usually there are separate mask patterns for each of the color separations. The contents of the mask patterns are bitwise ANDed with the incoming binary print image in order to gate and disable the jets from firing on the non-allowed locations. The mask patterns are convenient for implementation since they also allow the support of checker-boarding patterns often used with the various print modes and multiple number of passes that are common with these inkjet devices.
However, by restricting the black jets to fire on a fixed 300-dpi grid, the effective quality of text and other black features is greatly compromised. Furthermore, additional image processing is usually required in order to take into account the differences in ink-volumes and ensure the proper [neighborhood-wide] under-color removal. Often, the additional cost associated with these operations prohibits their use for the low-end high-volume inkjet devices. Instead, manufacturers often resort to relying on ink-reduction techniques, but these, too, tend to compromise quality by forcing a limit to the total amount of ink or utilizing the under-color removal for that purpose.
The current method of processing data for print heads is documented in
FIG. 1
which is an image processing block diagram. All text, line-art, and images are converted and treated as a single bitmap object prior to color correction and rendering. All the pixels comprising the print page are treated the same regardless of their source of origin (be it text, line art, or image). The following is a brief summary of the existing method.
Referring to
FIG. 1
, the RGB data is first passed through a gamma correction table to compensate for the difference between the device RGB and the RGB primaries used for generating the 3D color correction lookup table. It also takes into account the device non-linearity's and the different surrounding effects between the input and output. Since the device primaries and non-linearity's are impossible to obtain, they were replaced with the gamma correction. The gamma function values were empirically determined from 1.2 for plain paper to 1.8 for HQ glossy paper to 2.0 for transparency. The gamma corrected RGB was then color-space converted to YES, the Y component further compressed by passing through a TRC, and converted back to RGB. The modified RGB values were then used as input to the main 3D color-correction lookup table, where they were converted to CMY. A GCR/UCR (under color removal) function was used to generate the K component, which was then passed through four 1D linearizing TRCs and the final 1D ink-limit TRCs. The ink-limited CMYK data was then used to fire the jets in accordance with a pre-determined dot-scheduling algorithm that was selected to match the desired media type.
With respect to the existing color correction scheme, the UCR/GCR strategy for the print head is to replace the primary color inks (CMY) with black ink (K) at a level that minimizes the graininess. However, since black dots have nearly twice the drop volume and therefore are very visible, the approach taken was to not remove CMY for light colors (i.e., no addition of black ink) in order to eliminate the situation of grainy sparse black dots. For darker colors, on the other hand, the desire was to partially replace CMY with black ink.
The final recommendation for coated paper was to apply two different functions for the subtraction of CMY and the addition of K, both starting from a density level that optimizes the graininess (0.15 to 0.25). While the K addition function continues to grow monotonically, the CMY subtraction function reaches a peak and then falls back to zero. This algorithm has the advantage of no K in the highlights and good maximum solid area density (up to 400% ink), but because it puts down a lot of ink in dark colors it requires aggressive ink limitation TRCs (toner reproduction curve). More over, the algorithm fails to control the inks on plain paper, particularly for media with a severe show-through problem.
A more complicated function consisting of three parts: graininess control, parabolic section, and a linear part, may be used for this case. In addition, the maximum coverage of CMYK inks must be limited based on the amount of show-through, inter-color bleeding, and chrominance levels. Since the spot size of the black ink is twice as large as CMY, it is not sufficient to limit the ink in CMY. It was essential to use multiple passes in some of the driver print modes. For example, the High Quality printing mode at 600×600 dpi uses 4 passes for CMY and 8 passes for K, with 50% color coverage for CMY and 25% for K. Especially dark colors composed of CMYK values would tend to generate more ink than the media are capable of holding. Therefore the ink-limitation method is critical for ensuring optimum color gamut and dynamic range.
A major drawback of the existing method is that the resulting multi-pass dot-scheduling algorithm that was derived based on the above global considerations has made it very non-flexible with respect to the local neighborhood requirements. To illustrate this point, consider an example of a two-pass dot-scheduling algorithm in which only the odd dots are allowed to be exercised in the first pass, and then only the even dots in the second pass. Assume further that the media is limited in the amount of black ink it can take (which is almost always the case at 400% ink). The glo
Cunha Robert
Nguyen Thinh
Xerox Corporation
LandOfFree
Enhanced text and line-art quality for multi-resolution... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Enhanced text and line-art quality for multi-resolution..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Enhanced text and line-art quality for multi-resolution... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2516947