Facsimile and static presentation processing – Static presentation processing – Attribute control
Reexamination Certificate
1999-10-29
2001-02-06
Rogers, Scott (Department: 2724)
Facsimile and static presentation processing
Static presentation processing
Attribute control
C358S001200, C358S451000, C358S451000, C358S451000
Reexamination Certificate
active
06185008
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to laser printing systems, and more particularly to methods, apparatus, systems and media for producing accurate and visually-pleasing laser printer graphics.
2. Description of the Related Art
A laser printer generates an image on a piece of paper by scanning a Focused laser beam over a cylindrical photosensitive drum, the signal directed to the laser beam for each pixel controlling the length of time which the laser beam is on. The drum converts the laser power incident on the drum to electrostatic charges, and the electrostatic charges attract and retain a powdered ink or “toner” (generically referred to as a “marking medium”). When an electrostatically charged paper (generically referred to as a “printing surface”) is rolled against the drum, the toner is transferred to the paper, and the paper is then heated to fuse the toner to the paper. The resolution of the image is determined by the number of scan lines per inch (lpi) and the density of pixel values along each scan line, as measured in dots per inch (dpi). A typical high-resolution laser printer has 600 scan lines per inch and 600 dots per inch along each scan line.
A graph of an exemplary laser signal
100
as a function of pixel position is shown in
FIG. 1A
, and a graph of the resulting line thickness
130
and a magnified view of the toner line
170
generated by the laser printer are shown in
FIGS. 1B and 1C
, respectively. (For clarity, the toner line
170
is shown with no toner in the scan lines
160
and
180
directly above and below it
170
, respectively, although generally there will be toner in those scan lines
160
and
180
as well.) As may be noted by inspection of
FIGS. 1A-C
, the thickness of the toner line
170
is roughly proportional to the signal level directed to the laser. The thickness of the toner line
170
reaches a width of one pixel when the level of the laser signal has its full-pixel-width value, PP, and the thickness of the line
170
having a maximum toner width, WM, (which is three pixels in the example of
FIG. 1C
) when the level of the laser signal has its maximum value, PM.
While a monochromatic image will only require one toner line per scan line, a full-color image will require multiple colored inks per scan line. Most commonly, full-color printed images are generated using cyan, magenta, yellow and black inks, and the image is referred to as the “CMYK” image. Although the present invention applies to full-color image printing as well as monochromatic image printing, for ease of discussion of the present invention only one toner color per scan line is discussed, since the generalization to multiple toners per scan line is straightforward.
Although laser printers can generate images quickly, nonlinearities in the relation between the level directed to the laser and the amount of deposited toner are problematic in the accurate generation of images, especially in regions where the toner density is low (highlight regions) and regions where the toner density is high (shadow regions). One of the nonlinearities associated with the process is the result of interparticulate attraction between toner particles which causes them to clump together, preventing dots of toner deposited on the paper from being below a minimum size. Another cause of nonlinearities between the signal applied to the laser and the resulting toner density is a result of the time required to turn the laser off and on. As illustrated by the graph of the laser signal as a function of pixel number of FIG.
1
D and the corresponding graph of incident laser power as a function of distance on the toner drum of
FIG. 1E
, the length of time that the laser power signal has the maximum laser power value (which is normalized to a value of unity in
FIG. 1E
) is roughly proportional to the laser signal value. However, the laser power cannot reach the value of unity instantaneously, nor cannot it return to zero power instantaneously. These ramp times effect the proportionality between the laser signal and the laser power. The non-proportionality is most dramatic for small laser signals, such as the signal at pixel number
1
, where the finite-length rise and fall times of the laser power prevent the laser power from reaching the unity value.
A graph of an exemplary low-power laser signal
200
is shown in
FIG. 2A
, and a graph of the resulting line thickness
230
and a magnified view of the toner line
260
generated by the laser printer are shown in
FIGS. 2B and 2C
, respectively. As may be noted by inspection of
FIGS. 2A-C
, for pixels which are surrounded by laser values of zero and have laser values which are below an isolated-single-pixel cutoff value, PC
1
, such as the second pixel, the laser engine cannot deposit any toner on the paper. However, for a pixel which is surrounded by laser values of zero and is above the isolated-single-pixel cutoff value PC
1
(such as the fourth pixel), the laser engine can generate a toner covered area
262
. Similarly, the laser engine cannot generate a toner covered area for a pair of laser values which are surrounded by laser values of zero, and for which the average value is below an isolated-double-pixels cutoff value, PC
2
. However, when an isolated pair of pixels, such as the sixth and seventh pixels, have an average value above the isolated-double-pixels cutoff value PC
2
, a toner covered area
264
is produced. Similarly, the isolated-triple-pixels cutoff PC
3
has a value below the isolated-double-pixels cutoff value PC
2
, and when an isolated triple of pixels (not shown) have an average value above the isolated-triple-pixels cutoff value PC
3
, a toner covered area is produced, and so on.
The situation is actually somewhat more complex than indicated above, since non-zero values of surrounding pixels influence the cutoff values. For instance, when the laser values surrounding a pixel have a value of one, the smallest printable dot corresponds to a value somewhat less than PC
1
. And when the laser values surrounding the dot have a value of two, the smallest printable dot corresponds to a value somewhat less still. Similarly, when the laser values surrounding a pair of pixels have a value of one, the smallest printable dot corresponds to an average value of the pair of pixels somewhat less than PC
2
.
Another nonlinearity of laser printer imaging is found in the larger laser values which produce shadowed regions. A graph of an exemplary laser signal
300
is shown in
FIG. 3A
, and a graph of the resulting line thickness
330
and a magnified view of the toner line
360
generated by the laser printer are shown in
FIGS. 3B and 3C
, respectively. As may be noted by inspection of
FIGS. 3A-C
, for laser values
300
which are near the middle of the range between zero and the full-pixel-width value PP (in this case the first through sixth pixels), the laser engine generates a line
360
whose thickness is roughly proportional to the laser values. However, for laser values which are near the full-pixel-width value PP (such as those of the seventh, eighth and ninth pixels), the laser engine tends to overly darken the pixels.
The underemphasis of low-power laser signals and the overemphasis of high-power laser signals (in scan lines of uniform darkness) is represented by the graph
400
of
FIG. 4A
, which plots laser value versus toner line thickness, i.e., toner “density.” As can be seen from the graph
400
, for low values
410
of the laser signal the line thickness is zero, as discussed above in reference to
FIGS. 2A-C
, and for high values
430
of the laser signal the line thickness reaches unity prior to the laser signal reaching the value PP, as discussed above in reference to
FIGS. 3A-C
.
Another drawback illustrated by
FIG. 4A
, is the fact that the laser values have a finite number of discrete values and the line thicknesses therefore have a finite number of discrete thicknesses. This “quantization” of line thicknesses can produce “false contours” when there is an
Li Chia-Hsin
Pascovici Andrei
Shu Joseph
Rogers Scott
Seiko Epson Corporation
LandOfFree
Multilevel screening for color laser printer does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Multilevel screening for color laser printer, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multilevel screening for color laser printer will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2614138