Facsimile and static presentation processing – Static presentation processing – Attribute control
Reexamination Certificate
1999-01-11
2002-12-10
Lee, Thomas D. (Department: 2624)
Facsimile and static presentation processing
Static presentation processing
Attribute control
C358S003130
Reexamination Certificate
active
06493112
ABSTRACT:
BACKGROUND OF THE INVENTION
Field of the Invention
The present invention relates generally to half-toning of images, and, more particularly, to a method and apparatus for producing halftone images using green-noise masks having adjustable coarseness.
Description of the Related Art
Many color printing devices produce color output by printing dots of different colored inks. In these instances, each pixel of a color image is represented by a color vector which is typically a CMYK (cyan, magenta, yellow, black) quadruplet. Each element within the vector is referred to as a color component in the CMYK color vector. It is sometimes convenient to consider a color image, such as a CMYK image, as the overlapping of gray-scale images where there is one gray-scale image for any one, some or all color components, and the gray-scale image for any one particular color component is the amount of that color and not necessarily the amount of black within the original color image.
Most commercial devices such as, laser printers, ink-jet printers and facsimile machines, are, with respect to any particular color component, binary-level, that is, capable of printing or not printing dots of ink. Typically, such devices are incapable of reproducing gray-scale images with respect to any particular color component. In order to reproduce gray-scale images, with respect to any particular color component, having a multitude of gray-scale values, these devices must convert the gray-scale images to binary images having only two gray-scale levels, the color component or white. The conversion must be such that the overlapping resultant binary image, when viewed by the human eye, appears to have several gray-scale values, with respect to each color component, when in fact it is just a combination of that color and white dots. Methods to convert gray-scale images to binary images for printing are referred to as “halftoning.” As used herein, the term “gray-scale image” refers to the gray-scale image of a color image such as a CMYK image, as well as a black and white image.
Halftoning renders the illusion of various shades of gray by using only two levels, black and white, and can be implemented either digitally or optically. In the case of digital halftoning, points correspond to pixels. Many digital halftoning methods exist in practice today. Clustered-dot dithering, which is similar to the analog method used to render images in newspapers, and error diffusion are two such methods. The method known as halftoning via “blue-noise masks,” however, is the preferred method used in the reproduction industry due to its simplicity and performance. As referred to herein, the term “blue-noise” is a pattern having predominantly high frequency components of white noise and which possesses certain visually pleasing properties, as described in U.S. Pat. No. 5,111,310 to Parker et al. Other halftoning methods of similar complexity to blue-noise masks produce significantly inferior halftone images.
The concept of blue-noise and its associated spectral characteristics have had a profound impact in digital printing technology. R. Ulichney, in “Dithering With Blue-noise,”
Proceeding Of The IEEE
, January 1988, first observed that the spectral characteristics of homogeneous binary patterns created by error diffusing constant gray signals, closely follow the spectral characteristics of blue-noise and its high-frequency components. Ulichney studied the one-dimensional radially averaged power spectrum density (RAPSD) of a blue-noise binary pattern of a given gray-level g. The cut-off frequency f
6
, also known as the principle frequency, determines the average distance between the minority pixels of a binary blue-noise pattern. Principle frequency f
6
depends on the gray-level and is expressed as:
f
b
=
g
R
,
for
0
<
g
≤
0.5
1
-
g
R
,
for
0.5
<
g
≤
1.0
(
1
)
where R is the minimum distance between addressable points in the display. An additional measure utilized for analyzing binary dither patterns consists of the pair correlation defined by D. Stoyan, W. S. Kendall, and J. Mecke in
Stochiastic Geometry and Its Applications
, (John Wiley and Sons, New York, 1987) [hereinafter referred to as “Stoyan et al.”]. As defined by Stoyan et al., a point process &PHgr; is a stochastic model governing the location of points x
i
within the space R
2
, where R
2
is the two-dimensional real space. A sample &phgr; of the point process &PHgr; is written as &phgr;={x
i
&egr;R
2
: i=1, . . . , N}, and a scalar quantity &phgr;(B) is defined as the number of points x
i
in the subset B of R
2
. It is assumed that the point process &phgr; is simple, meaning that i≠j implies x
i
≠x
j
, which further implies:
lim
φ
(
dV
x
)
dV
x
→
0
=
{
1
for
x
∈
φ
0
else
(
2
)
where dV
x
is the infinitesimally small area around x. In terms of a discrete dither pattern, sample &phgr; represents the set of minority pixels such that &phgr;[n]=1 indicates a minority pixel at location n. A minority pixel is a pixel which is “on” while more than half of all pixels are “off,” and which is “off” when more than half of all pixels are “on.” A pixel is deemed to be turned “on” when the pixel has a smaller value than a predetermined gray level g value. A pixel is deemed to be turned “off” when the pixel has a larger value than the predetermined gray level g value. If exactly half of all pixels are “on,” then either group may be treated as the minority pixels.
Given &phgr;&egr;&PHgr;, a pair correlation R(r) is defined as:
R
(
r
)
=
E
(
φ
(
R
y
(
r
)
)
|
y
∈
φ
)
E
(
φ
(
R
y
(
r
)
)
)
(
3
)
where R
y
(r) specifies the ring centered around the point y&egr;&PHgr; with an inner radius r and an outer radius r+dr, as shown in FIG.
1
. In terms of a binary dither pattern, R(r) is the ratio of the average number of minority pixels located a distance d away from the minority pixel at sample y, such that r≦d<r+dr, to the average number of minority pixels in a region of size N
r
pixels with a gray-level g. Region size N
r
is the total number or pixels located a distance d away from the minority pixel located at sample y, such that r≦d<r+dr, and g is the average gray-level of the dither pattern.
As described in U.S. Pat. No. 5,111,310 issued to Parker et al., blue-noise mask halftoning, also referred to as frequency modulated (FM) screening, performs a pixel-by-pixel comparison of the gray-scale image against a halftone screen or mask having high, radially-isotropic frequency, i.e., blue-noise, characteristics. Under ideal printing conditions, blue-noise mask halftoning provides optimal rendering of gray-scale images when viewed by the human eye. However, in real printers using blue-noise mask halftoning, printed black dots are not perfect squares and neighboring pixels overlap each other causing printed images to appear darker than desired. In addition, printer distortions such as dot gain further degrade image quality. These image defects are called “artifacts” because they represent an artificial feature caused by the method of image creation instead of a true feature of the desired image.
A modified approach to the construction of a blue mask is reported in M. Yao and K. J. Parker, “Modified approach to the Construction of a blue-noise mask,”
Journal of Electronic Imaging
, Vol. 3, January 1994. A hybrid deterministic/random approach is used in Y. Meng and K. J. Parker, “Dot gain compensation in the blue-noise mask,” No. VI in
Human Vision, Visual Processing, and Digital Display VI
, (
SPIE
), 1995, for modification of blue-noise masks in order to compensate for the dot-gain problem present in printing hardware. Finally, a radially asymmetric FM mask (blue-noise mask) is derived using a novel optimization algorithm as noted in J. Allebach and Q. Lin, “FM screen design using dbs algorithm,” No. V
Arce Gonzalo R.
Lau Daniel L.
Brinich Stephen
Connolly Bove & Lodge & Hutz LLP
Lee Thomas D.
University of Delaware
LandOfFree
Method and apparatus for producing halftone images using... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for producing halftone images using..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for producing halftone images using... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2917787