Facsimile and static presentation processing – Static presentation processing – Size – resolution – or scale control
Reexamination Certificate
1999-11-04
2004-09-21
Rogers, Scott (Department: 2626)
Facsimile and static presentation processing
Static presentation processing
Size, resolution, or scale control
C358S003270
Reexamination Certificate
active
06795211
ABSTRACT:
TECHNICAL FEILD
The present invention relates generally to resolution enhancement technology (RET) and more particularly to methods and systems for enhancing resolution of compressed image data.
BACKGROUND OF THE INVENTION
Resolution Enhancement Technology (RET) is a well-established technology for the improvement of printed text. Resolution enhancement of input image data involves pattern-matching techniques to smooth text edges and avoid jaggedness. A large number of methods exist. In general, with RET techniques the edges of letters are recorded and diagonal edges are smoothed. A number of these RET techniques are summarized below.
U.S. Pat. No. 5,282,057 to Mailloux, et al. discloses a bit-map image resolution converter for converting binary image data originally generated at a lower resolution into representative binary image data of a higher resolution, wherein the conversion ratio, or magnification factor, is an integer value. This patent also discloses methods for smoothing the interpolated output image to thereby reduce objectionable visual characteristic that are observable in digitally encoded data using conventional magnification techniques.
U.S. Pat. No. 5,387,985 to Loce, et al. teaches a non-integer image resolution conversion using statistically generated look-up tables for converting the resolution of bitmap images, and using a template matching process to alter the resolution of digital images for printing or similar methods of rendition. Statistically generated templates, implemented using look-up tables, are used to improve document appearance by converting from an original input spatial resolution to an output spatial resolution that is device dependent, where there is a non-integer relationship between the input and output resolutions. The resulting image signals may then be utilized to control a scanning beam where the beam varies in intensity and duration according to the pulses used to control it.
U.S. Pat. No. 5,579,445 to Loce, et al. discloses a method and apparatus for automating the design of morphological or template-based filters for print quality enhancement. A plurality of different phase, but same resolution, subsampled images are generated from training documents. Statistical data derived therefrom is then employed in an automated process to generate filters. The filters may be used for resolution enhancement and/or conversion of bitmap images. Furthermore, the statistical data is used to produce filters that are intended to not only optimize image structure, but image density as well.
U.S. Pat. No. 5,696,845 to Loce, et al. teaches a method and apparatus for improving the appearance of printed documents, and more specifically, using a template matching process to enhance the fast-scan resolution of digital images while maintaining raster resolution for printing. Multiple-bit per pixel, statistically generated templates, implemented using look-up tables, are used to improve document appearance by converting from a single-bit per pixel to N-bits per pixel, while preserving raster resolution of the printed output. The resulting N-bit per pixel image signals may be utilized to control a scanning beam where the beam varies in intensity and duration according to the pulses used to control it.
U.S. Pat. No. 5,724,455 to Eschbach discloses an automated template design method for print enhancement for automating the design of pattern matching templates used to implement a print quality enhancement filter. More particularly, the method is directed to the use of a representative set of page description language decomposed training documents, and statistical data derived therefrom, in an automated process which generates templates for resolution enhancement and/or conversion of bitmap images.
The above described RET techniques use template-matching filters and are applied basically to bitmap pixel data not to compressed data. In resolution enhancement of input image data, the pixels are not only magnified by replication in each direction but a template filter is also applied to smooth edges and correct jaggedness. FIG. I is a simple block diagram in which the replication and template matching filtering are applied to the compressed data. If the image containing text is compressed with CCITT compression schemes, the enhancement involves the steps of decompressing the input image data, replicating the compressed data, applying template-matching filter to the replicated data, and recompressing the resolution enhanced pixel data.
As shown in
FIG. 1
, the compressed data is input and decompressed in the decompressing unit
101
. The decompressed data is applied to the replication unit
102
and the number of pixels is magnified in replication unit
102
to enhance the resolution of the input image. The replicated pixel data is sent to the template matching filter
103
and the edge of replicated pixel data is smoothed in the template matching filter
103
. The replicated and smoothed pixel data is compressed again in recompressing unit
104
. From this figure, a decompressing unit
301
and a recompressing unit
304
are needed to apply the template-matching filter to compressed data.
FIGS. 2A-2I
depict the bitmap examples of each output in FIG.
1
and each group of
FIGS. 2A-2C
,
2
D-
2
F and
2
G-
2
I illustrates an example of up-scaling as 2×2, 3×3 and 2×1, respectively.
FIG. 2A
is the output of decompressing unit
101
, which is decompressed bitmap pixel data.
FIG. 2B
is the output of replication unit
102
in which the bitmap of decompressed pixel data in
FIG. 2A
is replicated by 2×2.
FIG. 2C
is the output of template matching filter
103
in which the bitmap of pixel data up-scaled by 2×2 in
FIG. 2B
is edge smoothed.
FIG. 2D
is the output of decompressing unit
101
, which is decompressed bitmap pixel data.
FIG. 2E
is the output of replication unit
102
in which the bitmap of decompressed pixel data in
FIG. 2D
is replicated by 3×3.
FIG. 2F
is the output of template matching filter
103
in which the bitmap of pixel data up-scaled by 3×3 in
FIG. 2E
is edge smoothed.
FIG. 2G
is the output of decompressing unit
101
, which is decompressed bitmap pixel data.
FIG. 2H
is the output of replication unit
102
in which the bitmap of decompressed pixel data in
FIG. 2G
is replicated by 2×1.
FIG. 2I
is the output of template matching filter
103
in which the bitmap of pixel data up-scaled by 2×1 in
FIG. 2H
is edge smoothed.
In order to obtain the resolution enhanced pixel data, the device in
FIG. 1
needs amount of buffering and processing steps, which may cost up and be inappropriate in low-end level application.
SUMMARY OF THE INVENTION
The present invention provides methods and systems for increasing the resolution of image data. With the present invention all resolution enhancing technology (RET) operations may be performed in the compressed domain, so that the resolution enhancement is applied to the transition maps and not to pixel image data. The present invention provides a simple manipulation of the compressed image data, which effectively increases the image of resolution and smooth edges in the image, while avoiding full decompression and recompression of the data. This results in significant savings in time and computational overhead.
In accordance with one aspect of the present invention, a method is practiced in a compressed domain to enhance a resolution of input image data. For the purpose of the present invention, (a) the transitions of the input lines of the compressed image data are analyzed. (b) The positional differences between the closest compatible transitions in the input lines are calculated. (c) Distances of transitions from a reference position in each input line are magnified. (d) A line is interpolated between the input lines and the transitions of the interpolated line are determined according to the positional differences. (e) The corrected transitions of each line are encoded to compressed codes.
In accordance with another aspect of the pr
de Queiroz Ricardo L.
Eschbach Reiner
Fay Sharpe Fagan Minnich & McKee LLP
Rogers Scott
Xerox Corporation
LandOfFree
Resolution enhancement of compressed image data does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Resolution enhancement of compressed image data, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Resolution enhancement of compressed image data will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3264017