Facsimile and static presentation processing – Static presentation processing – Attribute control
Reexamination Certificate
1998-10-31
2002-08-27
Evans, Arthur G. (Department: 2622)
Facsimile and static presentation processing
Static presentation processing
Attribute control
C358S001150
Reexamination Certificate
active
06441922
ABSTRACT:
FIELD OF THE INVENTION
This invention relates generally to machines and procedures for incremental printing of text or graphics on printing media such as paper, transparency stock, or other glossy media; and most particularly to a scanning thermal-inkjet machine and method that construct images from individual ink spots created on a printing medium, in a two-dimensional pixel array. The invention is also believed applicable to sister technologies such as the hot wax transfer method. To optimize image quality the invention employs printmode techniques that are in some cases substantially randomized and in other cases arbitrary, but preferred embodiments of some facets of the invention invoke such printmode techniques through upstream tonal-level strategies.
BACKGROUND OF THE INVENTION
A basic goal for these procedures and machines is very high quality in printed images, using a relatively inexpensive printer. Incremental printing nowadays is generally accomplished through digital manipulation of image data in one or another type of electronic digital microprocessor.
All such manipulation, including the stages discussed below under the conventional designations of “image processing” and “printmasking”, can be performed in a host computer, e.g. in software that operates an attached printer, or can be built into the printer itself—but most commonly is shared between the two. In still other systems some of the manipulation is performed in yet another distinct product, such as for example a so-called “raster image processor” (RIP) to avoid overcommitting either the computer or the printer.
For operations performed within the printer or within an RIP, as is well known, the product may contain either a general-purpose digital processor running programs called “firmware”, or an application-specific integrated circuit (ASIC) manufactured to perform only specific functions of particular printers or RIPs. In some cases the printer or RIP may use both a firmware subsystem and an ASIC.
Image processing—The fundamental task of all these devices is receiving data representing a desired image and developing from those data specific moment-by-moment commands to a printing mechanism. This task, for purposes of the present document, will be called “image processing”.
Such processing typically includes, at the outset, some form of darkness and contrast control or adjustment. In a color printer, this preprocessing stage analogously also includes color conversions and any needed color corrections. For purposes of generality in the remainder of this document (except where otherwise clear from the context), the terms “color” and “colorimetric” will be used to encompass nonchromatic as well as chromatic colors, color specifications and the color manipulations just mentioned. Such preprocessing can handle both user-desired color modification and any known mismatch between an input-image color specification and the operating color space and gamut of the printer.
Next downstream from contrast, darkness and other color corrections—and particularly important for images other than text—image processing also includes rendering or rendition techniques (such as dithering of error diffusion). A rendition stage may be regarded as having two principal functions, both directed to making spatial assignments of color ink spots to particular pixels.
First, it attempts to implement the relatively continuous or very fine tonal gradations of a photograph-like image, in terms of the relatively limited number of gradations which a typical inexpensive printer can produce. A digital file in a computer ordinarily is able to represent fine tonal gradations quite accurately, since data formats—although digital—usually allow for at least 256 distinct tonal levels between, for instance, pure white and dead black.
Second, in a color printer, rendition also attempts analogously to implement the relatively huge number of chromatic colors which a computer can invoke. Rendition must accomplish this in terms of the relatively limited number of chromatic colors which a typical inexpensive printer can produce.
Applying the broader definition of “color” suggested four paragraphs above, these two functions essentially collapse into a single broader functional concept. In other words, in the technical parlance of color science for incremental printing, both these functions may be regarded as implementing complex multilevel “color” values, in a printing system that can directly produce only a very limited number of “color” values.
Banding—An obstacle to highest-quality printing is caused by repeating failure of particular elements of the print mechanisms to mark properly—or to mark consistently with other elements. Periodic artifacts arise from constant or repeating errors of inkjet trajectory, pen positioning and speed, and printing-medium positioning and speed.
For instance malfunction or misalignment of a particular inking nozzle or the like can leave a generally consistent white or light pixel row across every image region where that particular element (e.g. nozzle) is supposed to mark. In the case of misalignment, the same problem also produces excess inking across some nearby region where the same element should not be marking.
This very simple example is only meant as a basic introduction to the concept of banding. As will shortly be seen, banding encompasses patterned artifacts that are far, far more complicated, bewildering and difficult to trace, to comprehend or accordingly to eliminate.
Image regions are not all equally affected by such defects. The visual impact or significance of banding problems, or more generally of dot-placement errors, varies with the tonal level or in other words dot density within an image.
We can define three regions of a tonal ramp, based on the amount of white space:
(1) highlights: These areas have ample white space and to the naked eye exhibit little in the way of banding or other dot-placement artifacts.
Such artifacts are of course present, but hard to see—because small differences in dot position can represent only a relatively small fractional change (or none) in the large amount of white space that is seen. Furthermore, because the dots that are present are so far apart, and usually irregularly located, they fail to form a visual frame of reference within which a person can detect placement errors directly.
(2) midtones: These parts of the tonal range are most sensitive to banding because they have small amounts of white space in conjunction with moderate amounts of dot-filled space.
Dot-placement errors are highly visible because small differences in dot position can have a large effect on how much white space is visible—and in many situations also a disproportionate effect on the exact appearance of the moderate amount of dot-filled space. Coalescence contributes further to the conspicuousness of banding and graininess because dots clump together.
(3) saturated areas: These segments of the tonal range have almost no white space showing through, and again as in the highlights tend to exhibit minimal banding effects.
The large amount of colorant on the printing medium hides dot placement errors—with the exception of print-medium advance problems. Interactions between the colorant and the printing medium, however, can lead to flood banding and coalescence.
As a practical matter, the boundaries of these tonal-range segments depend in part upon the nature of the image being printed, as well as the exact character of the dot-placement errors produced by a particular printhead. Therefore these regions of the tonal ramp can be defined neither sharply nor generally.
As a rule of thumb, however, for purposes of placement-error visibility the midtone region has very roughly more than one single printed dot per four pixels—but, at the saturated end of the range, very roughly more than one single dot subtracted from full coverage, per four pixels. For example in a four-level (including zero) system, since the maximum number of dots in each pixel is three, the maximum inking in four pixels is 3×4&equ
Askeland Ronald A
Hoff Thomas S.
Hudson Kevin R
LandOfFree
Reduction of banding in incremental printing, through... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Reduction of banding in incremental printing, through..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Reduction of banding in incremental printing, through... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2905274