Facsimile and static presentation processing – Static presentation processing – Attribute control
Reexamination Certificate
1999-12-07
2003-09-02
Rogers, Scott (Department: 2624)
Facsimile and static presentation processing
Static presentation processing
Attribute control
C358S302000, C358S003010, C382S252000
Reexamination Certificate
active
06614557
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a method of transformation of images, and more especially, to a method for degrading the required bit number of grayscale images by an error diffusion scheme.
BACKGROUND OF THE INVENTION
In general, image-processing devices, such as scanners, printers, or fax machines, provide a certain number of memory bits to indicate a gray value of a picture element named pixel. For example, one bit is provided for a binary case where only two gray levels can be differentiated, typically, 1 for a spot and 0 for no spot. In another case, an 8-bit image-processing device provides 256 gray scales, where each gray value of a pixel is ranged from 0 to 255. As noted, the more bits can be provided for a pixel, the more gray levels can be differentiated.
Basically, different image-processing devices employ different resolutions established by different bit numbers. The resolution transformation is thus unavoidable when image data are transmitted from one device to another. For transformation from a higher resolution to a lower one, error diffusion is a well-known technique to provide such a transformation with a certain degree of degradation compensation.
When an image with a higher resolution is transformed to one with a lower resolution, image distortion will be caused due to portions of the bit number being truncated. To compensate for that, the truncated error of a pixel is distributed or diffused to some of its neighboring pixels, typically those nearest pixels downstream from the focus one.
FIG. 1A
shows a case in a focus pixel contributes its truncated error to the four nearest pixels downstream as indicated by the arrows.
FIG. 1B
shows for the same case that the truncated errors distributed in the focus pixel come from four nearest pixels upstream. Further, for the purpose of balancing image tone, the diffusions are preferably distributed in opposite directions along two immediate adjacent scan lines as shown in FIG.
1
C.
However, until now, for the error diffusion method, only the method for transforming a grayscale image to a binary image has been successfully developed. These binary cases can be easily achieved using a threshold value for comparison and truncation. In these cases, those original gray values greater than a predetermined threshold value are transformed to target values of 1, and those of a lower value are transformed to 0. However, there is no error diffusion method for transforming a grayscale image to one with reduced bit number instead of transforming to a binary image.
SUMMARY OF THE INVENTION
The present invention proposes an image transformation method to degrade a N-bit grayscale image to a reduced K-bit grayscale image. Error diffusion, error bias and a programmable integer C are utilized for the present method.
Firstly, for the focus pixel, the original grayscale value is read and denoted by G
i
, which is an N-bit integer with a value within [0, 2
N
−1]. The related errors E
j
previously obtained from some chosen pixels are subsequently loaded. The respective corresponding weighting coefficient W
j
for each error E
j
is determined. Then, a resultant gray value R
i
of the focus pixel after error diffusion can be calculated by the following equation:
R
i
=
G
i
+
∑
j
w
j
E
j
-
C
Herein, C is a programmable integer for compensating the error bias and, if necessary, for adjusting the brightness of the entire target image.
Thereafter, the degradation of the gray value can be implemented by dividing the gray value R
i
into a first part M
i
and a second part L
i
, wherein M
i
is a K-bit integer, L
i
is an (N−K)-bit integer, and M
i
and L
i
satisfy the following equations:
M
i
×2
N−K
+L
i
=R
i
−2
N−K
/2
≦L
i
<2
N−K
/2,
L
i
is then truncated form the resultant gray value R
i
and stored as the error of the focus pixel. The error E
i
of the focus pixel is stored as L
i
plus an error bias D of 2
N−K
/2, that is,
E
i
=L
i
+2
N−K
/2.
After the error bias D of 2
N−K
/2 is added, the computed errors will be stored as non-negative integers and fall in the range of 0≦E
i
<2
N−K
−1. Finally, the focus pixel is displayed using M
i
as the degraded target gray value.
REFERENCES:
patent: 5208871 (1993-05-01), Eschbach
patent: 5805738 (1998-09-01), Kaburagi et al.
patent: 6108105 (2000-08-01), Takeuchi et al.
Liu Ling-Yi
Lu Ai-Chieh
Yang Jazz
Destiny Technology Corporation
Dickinson Wright PLLC
Rogers Scott
LandOfFree
Method for degrading grayscale images using error-diffusion... 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 for degrading grayscale images using error-diffusion..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for degrading grayscale images using error-diffusion... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3024082