Facsimile and static presentation processing – Static presentation processing – Attribute control
Reexamination Certificate
1998-04-20
2002-09-10
Rogers, Scott (Department: 2624)
Facsimile and static presentation processing
Static presentation processing
Attribute control
C358S003260
Reexamination Certificate
active
06449061
ABSTRACT:
FIELD OF THE PRESENT INVENTION
The present invention is directed to a system and method for reducing hybrid error diffusion pattern shifting at certain grey levels and is dependent upon the image processing operations associated with the pixel being processed. More specifically, the present invention is directed to providing a dynamic noise profile to reduce hybrid error diffusion pattern shifting at certain grey levels by perturbing the threshold value used to binarize the image data wherein the noise profile is selected based on a window effect pointer or image classification of the pixel being processed and/or the grey level of the pixel being processed.
BACKGROUND OF THE PRESENT INVENTION
There are many methods of rendering grey images on an output device. One such example is error diffusion. Error diffusion can render complex images that contain a mixture of text and picture reasonably well. The utilization of error diffusion eliminates the need to have image segmentation to separate the text from the picture so that the picture aspect of the document can be screened and the text aspect of the document can be threshold.
FIG. 1
illustrates a flowchart of a typical error diffusion binarizafion system. In Step S
1
of this process, the video signal for pixel X is modified to include the accumulated error diffused to this pixel from previous threshold processes. The modified video signal value X is compared at Step S
2
with the value 128, assuming a video range between 0 and 255. If Step S
2
determines that the modified video signal value X is greater than or equal to 128, the process proceeds to Step S
4
wherein a value is output to indicate the turning ON of pixel X. The process then proceeds to calculate the error associated with the threshold process at Step S
6
wherein this error, Y, is calculate as being X−255.
On the other hand, if Step S
2
determines that the modified video signal value X is less than 128, a signal is output at Step S
3
indicating that the pixel X is to be turned OFF. The process then proceeds to Step S
5
wherein the error, Y, is calculated as being equal to the value X.
The error calculated in either Steps S
5
or S
6
is multiplied by weighting coefficients and distributed to downstream pixels in Step S
7
. Thus, the error from the threshold process is diffused to adjacent pixels. The coefficients conventionally used to diffuse the error to adjacent downstream pixels.
In addition to the typical error diffusion described above, hybrid high addressability error diffusion process can also be utilized which will be explained briefly.
Typically, the image processing architecture of a printing system uses either the functions of screening, thresholding, or error diffusion. For pixels to be screened, a similar modified video signal, V
S
′, is computed from the pixel video signal V and the screen value S at the pixel location. The modified video signal, V
S
′, for a conventional screening process is defined as V
S
′=(S+255−V)/2 in a system having 256 grey levels. The screen value S depends on the pixel location as well as the halftone screening pattern being used. It is noted that either a line screen or a dot screen can be used.
In the final step of binarization, the modified video signal, V
S
′, is compared with 128 to determine the ON or OFF characteristics of the pixel. Namely, if the modified video signal is greater than or equal to 128, the pixel should be OFF (black), otherwise it should be ON (white).
FIG. 2
illustrates a typical circuit for carrying the screening process wherein a screen value is added to the video signal by modulator
1
and comparator
3
compares the modified video signal with the threshold value. It is noted that this example gives the same result as the more typical approach of comparing the video V itself with a screen in lieu of the threshold value.
Hybrid error diffusion is the intertwining of the typical screening process with conventional error diffusion. Moreover, the typical error diffusion process can be extended to a high addressability environment. The blending of these three process will be discussed in more detail below.
To extend the conventional error diffusion process, described above, to a hybrid high addressability environment, the binarization (threshold) is performed at a higher spatial resolution, but the error computation and propagation is performed at the original lower spatial resolution. This splitting of the process substantially prevents or reduces the number of isolated subpixels, thereby maintaining high image quality.
In explaining the hybrid high addressability error diffusion process, it is assumed that the input grey levels at pixel location i and pixel location i+1 are represented by V
i
and V
i+1
, respectively, wherein V
i
′=(G
L
−V
i
)+(S
i
−Th), and V
i+1
′=(G
L
−V
i+1
)+(Si
i+1
−Th) wherein G
L
is the maximum grey level a pixel can have, S
i
and S
i+1
are the screen values for the pixels based on position of the pixels, and Th is the threshold value. The rendering error, at the lower resolution, that passes from upstream pixels to the downstream pixel location is denoted by e
i
.
It is noted that a feature of high addressability involves interpolation between pixels, the creation of subpixels. This interpolation impacts the hybrid high addressability error diffusion process. More specifically, depending on the way the interpolation is done, two distinct outputs can be obtained utilizing the high addressability error diffusion process. Each one of these distinct outputs will be discussed below.
With respect to a first interpolation scheme, the steps for determining the printing or rendering of a subpixel are as follows.
Initially, the modified pixel values P
0
i
=V
i
+e
i
and P
1
i
=V
i+l
+e
i
are computed wherein V
i
′=(G
L
−V
i
)+(S
i
−Th), and V
i+1
′=(G
L
−V
i+1
)+(S
i+1
−Th). The subpixels are denoted by 0 to N−1 wherein the high addressability characteristic is N. The high addressability characteristics is the number of subpixels that a printer can produce compared to the throughput bandwidth of the image processing system. In other words, the high addressability characteristic defined as the number of subpixels that the image output terminal can render from one pixel of image data.
High addressability is important in situations where the device can process the image data at one resolution, but print at a higher resolution. In such a situation, the present invention can take advantage of a processing system designed for a lower resolution image, (lower resolution can be processed quicker and less expensively), and a printing device which, through laser pulse manipulation, can print at a higher resolution. For example, the image can be processed at 600×600×8 and printed at 2400×600×1 using the high addressability process. In the above example, the high addressability characteristic is 4. If the image was processed at 600×600×8 and printed at 1200×600×1, the high addressability characteristic would be 2.
The interpolated subpixel values are computed as B
n
=P
0
+n(P
1
−P
0
)/N for n=0 to N−1. The interpolated subpixel values are then compared with a threshold value which in most cases is 128, assuming that the video value ranges from 0 to 255 (G
L
is equal to 255). If B
n
is greater than or equal to 128, the subpixel is turned ON; otherwise, the subpixel is turned OFF. The error to be propagated to downstream pixels is computed as the desired output, (P
0
+P
1
)/2, minus the actual output, namely, y*255/N, wherein y is the number of subpixels turned ON. The error is then multiplied by a set of weighting coefficients and distributed to the downstream pixels as in the first version.
More specifically, the screened inputted modified video signa
Eipert William
Nickerson Michael J.
Rogers Scott
Xerox Corporation
LandOfFree
System and method for providing dynamic noise profile... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for providing dynamic noise profile..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for providing dynamic noise profile... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2853508