CA1248878A - System for reproducing multi-level digital images on a bi-level printer of fixed dot size - Google Patents
System for reproducing multi-level digital images on a bi-level printer of fixed dot sizeInfo
- Publication number
- CA1248878A CA1248878A CA000502705A CA502705A CA1248878A CA 1248878 A CA1248878 A CA 1248878A CA 000502705 A CA000502705 A CA 000502705A CA 502705 A CA502705 A CA 502705A CA 1248878 A CA1248878 A CA 1248878A
- Authority
- CA
- Canada
- Prior art keywords
- gray scale
- block
- error
- scale values
- pixel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/40—Picture signal circuits
- H04N1/405—Halftoning, i.e. converting the picture signal of a continuous-tone original into a corresponding signal showing only two levels
- H04N1/4051—Halftoning, i.e. converting the picture signal of a continuous-tone original into a corresponding signal showing only two levels producing a dispersed dots halftone pattern, the dots having substantially the same size
- H04N1/4052—Halftoning, i.e. converting the picture signal of a continuous-tone original into a corresponding signal showing only two levels producing a dispersed dots halftone pattern, the dots having substantially the same size by error diffusion, i.e. transferring the binarising error to neighbouring dot decisions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/40—Picture signal circuits
- H04N1/405—Halftoning, i.e. converting the picture signal of a continuous-tone original into a corresponding signal showing only two levels
- H04N1/4055—Halftoning, i.e. converting the picture signal of a continuous-tone original into a corresponding signal showing only two levels producing a clustered dots or a size modulated halftone pattern
Abstract
ABSTRACT OF THE DISCLOSURE
A system for producing improved halftone images from continuous tone input images, the pixels of which have been digitized as gray scale values, using a printer having the capability to print a set of output pel patterns producing a respective set of discrete gray scale values, is disclosed.
Blocks of pel patterns with discrete gray scale values approximating the gray scale values of the input pixels are selected for printing, and the error in a first block due to the difference between the gray scale value associated with the block and the gray scale value of the corresponding input pixel, is determined. To diffuse this error it is distributed to at least two adjacent blocks using a random number generator to determine the dis-tribution such that a random fraction of said error is distributed to one block and the remainder is distributed to the other block. This error diffusion process is carried through the adjacent blocks taking into account the error fraction already as-signed. For edge enhancement the brightness gradient surrounding each input pixel is deter-mined and a threshold value is set below which said gradient is classified as small and above which it is classified as large. When said gradient is small, the selected discrete pattern is printed for the pixel, and when said gradient is large, the pel pattern is altered and the block representing the pixel is printed biasing the pels therein toward the darker input pixels.
A system for producing improved halftone images from continuous tone input images, the pixels of which have been digitized as gray scale values, using a printer having the capability to print a set of output pel patterns producing a respective set of discrete gray scale values, is disclosed.
Blocks of pel patterns with discrete gray scale values approximating the gray scale values of the input pixels are selected for printing, and the error in a first block due to the difference between the gray scale value associated with the block and the gray scale value of the corresponding input pixel, is determined. To diffuse this error it is distributed to at least two adjacent blocks using a random number generator to determine the dis-tribution such that a random fraction of said error is distributed to one block and the remainder is distributed to the other block. This error diffusion process is carried through the adjacent blocks taking into account the error fraction already as-signed. For edge enhancement the brightness gradient surrounding each input pixel is deter-mined and a threshold value is set below which said gradient is classified as small and above which it is classified as large. When said gradient is small, the selected discrete pattern is printed for the pixel, and when said gradient is large, the pel pattern is altered and the block representing the pixel is printed biasing the pels therein toward the darker input pixels.
Description
SYSTEM FOR REPRODUCING MULTi-LEVEL DIGITAL
IMAGES ON A
Bl-LEVEL PRINTER OF FIXED DOT SIZE
BACKGROUND OF THE INVENTION
The present invention is concerned with digital halftone generation techniques for the printing of digitized con-tinuous tone images on printers which have fixed print positions and a limited range of discrete print intensi-ties, the most common case being bi-level prlnters.
1~ In particular, the invention is directed to the utilization of error propagation to compensate for the absence of continuous printing of gray values, and an improved technique for edge enhancement.
In the conventional printing process, continuous tone images are prepared for printing by xposure through ~;
a halftone screen. ~ The ;resultaot ;image ~will have dots of various sizes, one for esch opening in~tbe screen.
These dots are iarger where more light penetrates the screen and smaller~whe~re~less light is~ present. Thus the exposure through the~s;creen converts :the g`ray yog8S-012~--T~ D : ~
:
.
~2~ 78 value of the original image at the opening in the screen into a dot of appropriate diameter.
In the digital analog of the above process, the original image is scanned with a scanner and the data are cap-S tured in a digital computer. These data consist of bi-nary representations of the scanned sray values; each such value ( typicaliy ranging from 0 to 255 ) corre-sponds to one scanned area or pixel of the input image.
To reproduce this image on a pririter capable of only printing black dots or pels in fixed output positions, it is necessary to create a sensation of gray by suitable choice of the pels to be printed. A summary of various techniques that have been used in the past is contained in the IBM Journal of Research and Development, tS Vol.26, No.6, pps.687-697. ~ ~ `
The scanned image may be scaled in the computer, but in any event, it may be assumed, without loss of gen-erality, that there is one pixel of the input image for each dot to be represented in the output. The~dots :
are represented in the outp~ut of a digital printer as patterns of pels, and the pel patterns~may be considered to be arranged in blocks. Fach pel in ~a biock is either printed or not by the~digltal printer.~ ;Each~ such block yog85-012 :
- , . ~
87~i3 represents the digital analog of the halftone dot used in conventional printing. Such a block may be of any shape which contains an integral number of pel positions and which, when replicated, fills the area to ~e printed without overlap. Square or rectangular block shapes are convenient. An extreme case is that where the block contains only one pel. Different levels of gray are ob-tained by printing or reproducing different pel patterns within a block. If the block contains only one pel, only two values of gray are availab!e. Additional pels within a block will permit the printing of additional discrete gray values. Larger blocks will give more discrete gray values or levels, but also tend to reduce the resolution of the output image.
lS An error occurs in the gray values of the output image when the gray levels available for printing within a block differ from the gray level determined for the re-spective block, which level is based on the brightness of the corresponding analogous halftone dot. Hence, the limited number of gray values available from the~
blocks limits the quality of the printed image~ due to the absence of correct grays in~ some oreas~ and also to~ the creation of contours as ~the reproduction process goes Y0985-01~ 3 ~
: , : :
"., : : ~ ~ :
. .:
from one block pattern to another in a region of slowly varying gray value.
These problems may be alleviated by intermixing the levels in different blocks in such a manner as to obtain the desired gray value as an average over many neighbouring blocks. A technique of "error propagation" used to achieve such an intermixing is described, for example, in U.S. Patent No. 4,668,995, issued May 26, 1987, and assigned to the same assignee as the present invention. In error techniques of this type, the gray value desired at a given block position in an output image is determined from the input gray level at that position added to the error propagated from previous positions. More particularly, to begin with, an output pel pattern for a given block is selected ~rom among the available patterns, which pattern typically represents an appoximation to the gray value desired for that block. The error is determined as the difference between the desired gray and the gray now associated with the given block. The determined error is then distrlbuted or assigned to one -, or more adjacent block positions on the same or subsequent lines of the image. This selection and error distribution method is then continued throughout the lines of the image taking~into :
.
3&7~3 account the assigned error when selecting the appro-priate pel pattern for each block.
A variety of methods hav~ been previously used for allocating thi~ distribution of the error. However, all divide the error into fixed proportions and thus have suffered to a greater or lesser extent from the problem of producing "worms", that is, output pattern artifacts consisting of apparent curved or straight lines.
In addition to the "worms", these methods tend to give fuzzy edges between dark and light regions of the im-age.
The present inve-ntion concerns itself with the elimi-nation of the "worms" and with the sharpening of the fuzzy edges.
::
~ SUMMARY OF THE INVENTION
It is accordingly an object of th~e present invention~to provide a system for achieving improved error~propa-gation and also for accomplishing edge enhancement.
: ~S :
In the prior art, as noted above, error diffusion is usually carried out by propagating the error at a par-ticular blocl< in fixed proportions to two or more adja-cent bloc~ positions. For example, the propagation of an error "e" to each of two blocks "a" and "b" might be accomplished using a relationship such as ea=0.5 e eb=0.5 e to calculate the error proportions propagated to blocks a and b. By comparison, in accordance with the present invention the error proportion is varied at random, with the aid of a source of pseudo random numbers. More particularly, in one implementation of the present in-vention the relationship to be used becomes, with "r"
denoting a pseudo-random number equidistributed on ~5 the range 0 to 1, ea=r e eb=e-ea wherein the sum of the propagated errors ea,eb equals the total error e. This implementation has been found to eliminate the "worms" noted in images reproduced with prior art processes. However, in some portions o~ some images a "pebbly" appearance may occur, in which event a more pleasing result may be obtained using the relationship ea= (0 . 25~0. 5 r) e eb=e-ea .
:
~LZ~ 8 Other random distributions of the error to be propa-gated may also be used as will be found appropriate.
In addition to, and in keeping with, the foregoing technique the present invention comprehends a method for enhancing the sharpness of the edges of objects occurring within the image. The edge enhancement technique of the present invention uses output blocks of pels and the associated brightness gradients, deter-mined from the input image. For example, the system may be constructed to operate when the output blocks are 3 x 3 pels or 3 x 6 pels. In the latter case, edge enhancement is carried out for each 3 x 3 half block independently. More particularly, the brightness gra-dient for each 3 x 3 pel region is firstly determined.
A threshold value is chosen below which the brightness gradient is classified as small and above which it is classified as large. When the gradient is small, the edge enhancement method is not used. When the gradient is large, the selection of the pel pattern approximating the gray value previously determined for the block or half block containing the region is discarded. Rather the appropriate number of the nine pels in the region to be printed is determined from the input gray value.
Associated with and corresponding to the output pels :
~5 YO985-01 2 rl ~ ~
'` ' ~
.
~g~878 in this 3 x 3 region are the input gray value for the region and the input gray values for the eight blocks surrounding the region. The largest gray value among these nine input gray values is found. The output pel in the 3 x 3 region corresponding in position to that of the block with this largest gray value, is then se-lected for printing. The block with the next largest gray value is found and the pel in the 3 x 3 region corresponding in position is then printed. This process is continued until the appropriate number of the nine pels, as previously determined, are selected for print-ing. As a result, the black or printed pels in the 3 x 3 region will be placed toward the neighboring regions with the darkest input valuesi thus yielding sharp edges.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a block diagram of a basic system for carrying :
out the present invention. ; ~ ~ ~
:
FIG. 2 illustrates block patterns ~used~ln printing, as prepared for a particular printer, in ~keeping with the ~
present invention. The ~first~column ~under the heading g :: :
:
~ :` :
~: :
~2g~
PATTERN shows individual biocks of 3 x 6 pels con-taining 17 different pel patterns, and the the second column shows corresponding areas containing nine identical blocks as they are arranged in printing. The S REFLECTIVITY column shows measured values of reflectivity (%) for a large area filled with replicas of the corresponding block. The GRAY VALUE column is derived from the reflectivities by a linear transformation and contains the gray values associated with the re-lt) spective blocks.
FIG. 3 illustrates the various blocks and half blocks of an output image. The direction of.propagation of error is indicated, for one block, by diagonal arrows.
FIG. 4 shows the relation of input pixels to output pels for edge enhancement.
FIG. 5 shows the placement of the output pels within a block to achieve edge enhancement~
~ . : :
DESCR~IPTION OF THE PREFERRED~ EMBODIMENT
YO985-012 ~ q ~
~L2~ 8 A basic system for carrying out the present invention is shown in FIG.1 and generally is composed of a means 1, e.g., an image scanner, for digitizing an original continuous tone image, such as a photograph 2, and providing it to a general purpose computer 3 wherein it is processed for printout as a halftone image on a digital printer 4.
An input image of the type to be acted upon by the present invention may be represerited by a series of gray values arranged in an array of L lines, each line containing I gray values. The gray values are typically expressed as integers in the range from 0 to 255. The output image is considered to consist of blocks of pels, each pel corresponding to an output dot or element that is printed by a digital printer or that otherwise makes up a tangible representation of the image. As seen in FIGS.2 and 3, the blocks to be described are composed of 3 x 6 pels for each pair of input gray valùes, so that there will by 9 times as many output pels as there are `input gray values. Referring to FIG. 2, 17 available block patterns are shown along with their associated reflectivities and gray values for a particular printer.
The output image will consis~ of lines of suitably chosen blocks from among the ~17 to ap~proxlmate the desired ~ ' ~2~ 7~
gray values in the various image areas. It will be seen that in alternate lines of the image, the output blocks are shifted by three pels when printed, corresponding to the conventional 45 degree halftone grid used in an-alog printing.
Referring next to FIG.3, the steps of the method or algorithm for carrying out error diffusion in accordance with the present invention will be described. The posi-tions on the input image are represented by a pair of numbers (I,i) where I is the line number and i is the number of the gray value in that line. The gray value for "a", the (I,i) pixel, in the input image is denoted gv(l,i). The gray value associated with a pattern number b will be designated g(b). A temporary array is used to store the errors, e(l,i).
The steps of the algorithm are as follows:
Initialize the errors for the first line: ~e.g., for i=1, ::
3, . . . , I-1 e(1, i)=0 Repeat the following sequence for all lines in the image.
For odd lines, the caiculation~ is ~carried ~out for l=1, 3, ,;
: :~
:
- ~ . - " . . ~, ~24~87'8 ..., I; for even lines for i=2, 4, ..., I. Initialize the error at the extreme left of the next line:
for an odd line e(l+1,0)=0 for an even lir.e e(l+1,1)=0 S Calculate the desired gray value d as the average of the gray values corresponding to the block:
d=(gv(l,i)+gv(l,i+1))/2 Find the trial block number b such -that d is equal to or greater than the gray value of block b and less than the gray yalue of block b+1. Select b such that g(bj=' d ' g(b+1) If d+e(l,i)=> g(b+1) increase b by 1. If b > 16, set b to 16.
:`
Place block b in the appropriate location on the output image, ready for printing.
;
Calculate and propagate the error x. First calculate t h e e r ro r x=e ( l, i ) +d -g ( b ) .
Distribute the error in~ accordance ~with ~the arrows in F I G . 3 .
YO985-(~1 2 , ~2~
e(l+1 ,i+1)=(0.25+0.5 r)x e(l+1,i-1)=e(1+1,i-1)+x-e(1+1,i+1) where r is a pseudo-randonl number equidistributed on the range 0 to 1.
Following distribution of the errors throughout the im-age in this manner, the output image will print out without worming or pebbling.
To perform edge enhancement for each half of the output block, if required, the following steps may be carried out( see FIG.4 ~:
.
.
For the left half, j=i; for the right half j-i+1.
A quantity proportional to the square of the magnitude of the intensity gradient is ~iven by : , ~
(gv(1+1 ,j+1 )-g\t(l-l ,j-1 ~)**2+(gv(1+1 , j~ (gv(l-l ,i+1)**2:
:
YO985-01~ ~ ~
:
1 3:
, ..
:
. ~ .
~ z~
If this quantity is greater than some threshold (e. g.
10,000), the corresponding half of the output block is modified to produce edge enhancement.
The number of black output pels n for the modified 3 x 3 half block is calculated as, e. 9., n=9 gv ( I, j ) /256 where n is rounded to the nearest integer.
Then, referring to FIG. 4, the n largest gray values from the set of FIG. 4 are found and the corresponding output pels in the 3 x 3 half block are made black or printed on the output image as shown in FIG.5.
Upon the application of both correction techniques de-scribed above an halftone output image may be obtained from a continous tone input image with clearly defined edges on the objects therein and without artifacts ;such as worming or pebbling in the other areas.
- , : :
T he system herein~ desc;ri~b~ed will produce ~printed~ images the same size as the scanned images on printers having three times the linear resolution of the scanner. For ')O YO985-01 2 :
:
example, a scanned image of 200 pels/inch may be printed out on a 600 pels/inch printer. A preferred printer for implementing the invention is the IBM 4250 which has been found to produce excellent electroerosion reproductions suitable for use in the preparation of long life lithographic plates. Various image scanners and other conventional means may be used in the process of capturing the images in a general purpose computer such as an IBM System 370 and the described algorithms 11) may be run by the computer. When it is necessary to change the output image size, the input image may be modified in size by any of a variety of algorithms suit-able for handling digital muiti-level images that are known in the art.
; .
1~ :
~:
: :
: : `
IMAGES ON A
Bl-LEVEL PRINTER OF FIXED DOT SIZE
BACKGROUND OF THE INVENTION
The present invention is concerned with digital halftone generation techniques for the printing of digitized con-tinuous tone images on printers which have fixed print positions and a limited range of discrete print intensi-ties, the most common case being bi-level prlnters.
1~ In particular, the invention is directed to the utilization of error propagation to compensate for the absence of continuous printing of gray values, and an improved technique for edge enhancement.
In the conventional printing process, continuous tone images are prepared for printing by xposure through ~;
a halftone screen. ~ The ;resultaot ;image ~will have dots of various sizes, one for esch opening in~tbe screen.
These dots are iarger where more light penetrates the screen and smaller~whe~re~less light is~ present. Thus the exposure through the~s;creen converts :the g`ray yog8S-012~--T~ D : ~
:
.
~2~ 78 value of the original image at the opening in the screen into a dot of appropriate diameter.
In the digital analog of the above process, the original image is scanned with a scanner and the data are cap-S tured in a digital computer. These data consist of bi-nary representations of the scanned sray values; each such value ( typicaliy ranging from 0 to 255 ) corre-sponds to one scanned area or pixel of the input image.
To reproduce this image on a pririter capable of only printing black dots or pels in fixed output positions, it is necessary to create a sensation of gray by suitable choice of the pels to be printed. A summary of various techniques that have been used in the past is contained in the IBM Journal of Research and Development, tS Vol.26, No.6, pps.687-697. ~ ~ `
The scanned image may be scaled in the computer, but in any event, it may be assumed, without loss of gen-erality, that there is one pixel of the input image for each dot to be represented in the output. The~dots :
are represented in the outp~ut of a digital printer as patterns of pels, and the pel patterns~may be considered to be arranged in blocks. Fach pel in ~a biock is either printed or not by the~digltal printer.~ ;Each~ such block yog85-012 :
- , . ~
87~i3 represents the digital analog of the halftone dot used in conventional printing. Such a block may be of any shape which contains an integral number of pel positions and which, when replicated, fills the area to ~e printed without overlap. Square or rectangular block shapes are convenient. An extreme case is that where the block contains only one pel. Different levels of gray are ob-tained by printing or reproducing different pel patterns within a block. If the block contains only one pel, only two values of gray are availab!e. Additional pels within a block will permit the printing of additional discrete gray values. Larger blocks will give more discrete gray values or levels, but also tend to reduce the resolution of the output image.
lS An error occurs in the gray values of the output image when the gray levels available for printing within a block differ from the gray level determined for the re-spective block, which level is based on the brightness of the corresponding analogous halftone dot. Hence, the limited number of gray values available from the~
blocks limits the quality of the printed image~ due to the absence of correct grays in~ some oreas~ and also to~ the creation of contours as ~the reproduction process goes Y0985-01~ 3 ~
: , : :
"., : : ~ ~ :
. .:
from one block pattern to another in a region of slowly varying gray value.
These problems may be alleviated by intermixing the levels in different blocks in such a manner as to obtain the desired gray value as an average over many neighbouring blocks. A technique of "error propagation" used to achieve such an intermixing is described, for example, in U.S. Patent No. 4,668,995, issued May 26, 1987, and assigned to the same assignee as the present invention. In error techniques of this type, the gray value desired at a given block position in an output image is determined from the input gray level at that position added to the error propagated from previous positions. More particularly, to begin with, an output pel pattern for a given block is selected ~rom among the available patterns, which pattern typically represents an appoximation to the gray value desired for that block. The error is determined as the difference between the desired gray and the gray now associated with the given block. The determined error is then distrlbuted or assigned to one -, or more adjacent block positions on the same or subsequent lines of the image. This selection and error distribution method is then continued throughout the lines of the image taking~into :
.
3&7~3 account the assigned error when selecting the appro-priate pel pattern for each block.
A variety of methods hav~ been previously used for allocating thi~ distribution of the error. However, all divide the error into fixed proportions and thus have suffered to a greater or lesser extent from the problem of producing "worms", that is, output pattern artifacts consisting of apparent curved or straight lines.
In addition to the "worms", these methods tend to give fuzzy edges between dark and light regions of the im-age.
The present inve-ntion concerns itself with the elimi-nation of the "worms" and with the sharpening of the fuzzy edges.
::
~ SUMMARY OF THE INVENTION
It is accordingly an object of th~e present invention~to provide a system for achieving improved error~propa-gation and also for accomplishing edge enhancement.
: ~S :
In the prior art, as noted above, error diffusion is usually carried out by propagating the error at a par-ticular blocl< in fixed proportions to two or more adja-cent bloc~ positions. For example, the propagation of an error "e" to each of two blocks "a" and "b" might be accomplished using a relationship such as ea=0.5 e eb=0.5 e to calculate the error proportions propagated to blocks a and b. By comparison, in accordance with the present invention the error proportion is varied at random, with the aid of a source of pseudo random numbers. More particularly, in one implementation of the present in-vention the relationship to be used becomes, with "r"
denoting a pseudo-random number equidistributed on ~5 the range 0 to 1, ea=r e eb=e-ea wherein the sum of the propagated errors ea,eb equals the total error e. This implementation has been found to eliminate the "worms" noted in images reproduced with prior art processes. However, in some portions o~ some images a "pebbly" appearance may occur, in which event a more pleasing result may be obtained using the relationship ea= (0 . 25~0. 5 r) e eb=e-ea .
:
~LZ~ 8 Other random distributions of the error to be propa-gated may also be used as will be found appropriate.
In addition to, and in keeping with, the foregoing technique the present invention comprehends a method for enhancing the sharpness of the edges of objects occurring within the image. The edge enhancement technique of the present invention uses output blocks of pels and the associated brightness gradients, deter-mined from the input image. For example, the system may be constructed to operate when the output blocks are 3 x 3 pels or 3 x 6 pels. In the latter case, edge enhancement is carried out for each 3 x 3 half block independently. More particularly, the brightness gra-dient for each 3 x 3 pel region is firstly determined.
A threshold value is chosen below which the brightness gradient is classified as small and above which it is classified as large. When the gradient is small, the edge enhancement method is not used. When the gradient is large, the selection of the pel pattern approximating the gray value previously determined for the block or half block containing the region is discarded. Rather the appropriate number of the nine pels in the region to be printed is determined from the input gray value.
Associated with and corresponding to the output pels :
~5 YO985-01 2 rl ~ ~
'` ' ~
.
~g~878 in this 3 x 3 region are the input gray value for the region and the input gray values for the eight blocks surrounding the region. The largest gray value among these nine input gray values is found. The output pel in the 3 x 3 region corresponding in position to that of the block with this largest gray value, is then se-lected for printing. The block with the next largest gray value is found and the pel in the 3 x 3 region corresponding in position is then printed. This process is continued until the appropriate number of the nine pels, as previously determined, are selected for print-ing. As a result, the black or printed pels in the 3 x 3 region will be placed toward the neighboring regions with the darkest input valuesi thus yielding sharp edges.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a block diagram of a basic system for carrying :
out the present invention. ; ~ ~ ~
:
FIG. 2 illustrates block patterns ~used~ln printing, as prepared for a particular printer, in ~keeping with the ~
present invention. The ~first~column ~under the heading g :: :
:
~ :` :
~: :
~2g~
PATTERN shows individual biocks of 3 x 6 pels con-taining 17 different pel patterns, and the the second column shows corresponding areas containing nine identical blocks as they are arranged in printing. The S REFLECTIVITY column shows measured values of reflectivity (%) for a large area filled with replicas of the corresponding block. The GRAY VALUE column is derived from the reflectivities by a linear transformation and contains the gray values associated with the re-lt) spective blocks.
FIG. 3 illustrates the various blocks and half blocks of an output image. The direction of.propagation of error is indicated, for one block, by diagonal arrows.
FIG. 4 shows the relation of input pixels to output pels for edge enhancement.
FIG. 5 shows the placement of the output pels within a block to achieve edge enhancement~
~ . : :
DESCR~IPTION OF THE PREFERRED~ EMBODIMENT
YO985-012 ~ q ~
~L2~ 8 A basic system for carrying out the present invention is shown in FIG.1 and generally is composed of a means 1, e.g., an image scanner, for digitizing an original continuous tone image, such as a photograph 2, and providing it to a general purpose computer 3 wherein it is processed for printout as a halftone image on a digital printer 4.
An input image of the type to be acted upon by the present invention may be represerited by a series of gray values arranged in an array of L lines, each line containing I gray values. The gray values are typically expressed as integers in the range from 0 to 255. The output image is considered to consist of blocks of pels, each pel corresponding to an output dot or element that is printed by a digital printer or that otherwise makes up a tangible representation of the image. As seen in FIGS.2 and 3, the blocks to be described are composed of 3 x 6 pels for each pair of input gray valùes, so that there will by 9 times as many output pels as there are `input gray values. Referring to FIG. 2, 17 available block patterns are shown along with their associated reflectivities and gray values for a particular printer.
The output image will consis~ of lines of suitably chosen blocks from among the ~17 to ap~proxlmate the desired ~ ' ~2~ 7~
gray values in the various image areas. It will be seen that in alternate lines of the image, the output blocks are shifted by three pels when printed, corresponding to the conventional 45 degree halftone grid used in an-alog printing.
Referring next to FIG.3, the steps of the method or algorithm for carrying out error diffusion in accordance with the present invention will be described. The posi-tions on the input image are represented by a pair of numbers (I,i) where I is the line number and i is the number of the gray value in that line. The gray value for "a", the (I,i) pixel, in the input image is denoted gv(l,i). The gray value associated with a pattern number b will be designated g(b). A temporary array is used to store the errors, e(l,i).
The steps of the algorithm are as follows:
Initialize the errors for the first line: ~e.g., for i=1, ::
3, . . . , I-1 e(1, i)=0 Repeat the following sequence for all lines in the image.
For odd lines, the caiculation~ is ~carried ~out for l=1, 3, ,;
: :~
:
- ~ . - " . . ~, ~24~87'8 ..., I; for even lines for i=2, 4, ..., I. Initialize the error at the extreme left of the next line:
for an odd line e(l+1,0)=0 for an even lir.e e(l+1,1)=0 S Calculate the desired gray value d as the average of the gray values corresponding to the block:
d=(gv(l,i)+gv(l,i+1))/2 Find the trial block number b such -that d is equal to or greater than the gray value of block b and less than the gray yalue of block b+1. Select b such that g(bj=' d ' g(b+1) If d+e(l,i)=> g(b+1) increase b by 1. If b > 16, set b to 16.
:`
Place block b in the appropriate location on the output image, ready for printing.
;
Calculate and propagate the error x. First calculate t h e e r ro r x=e ( l, i ) +d -g ( b ) .
Distribute the error in~ accordance ~with ~the arrows in F I G . 3 .
YO985-(~1 2 , ~2~
e(l+1 ,i+1)=(0.25+0.5 r)x e(l+1,i-1)=e(1+1,i-1)+x-e(1+1,i+1) where r is a pseudo-randonl number equidistributed on the range 0 to 1.
Following distribution of the errors throughout the im-age in this manner, the output image will print out without worming or pebbling.
To perform edge enhancement for each half of the output block, if required, the following steps may be carried out( see FIG.4 ~:
.
.
For the left half, j=i; for the right half j-i+1.
A quantity proportional to the square of the magnitude of the intensity gradient is ~iven by : , ~
(gv(1+1 ,j+1 )-g\t(l-l ,j-1 ~)**2+(gv(1+1 , j~ (gv(l-l ,i+1)**2:
:
YO985-01~ ~ ~
:
1 3:
, ..
:
. ~ .
~ z~
If this quantity is greater than some threshold (e. g.
10,000), the corresponding half of the output block is modified to produce edge enhancement.
The number of black output pels n for the modified 3 x 3 half block is calculated as, e. 9., n=9 gv ( I, j ) /256 where n is rounded to the nearest integer.
Then, referring to FIG. 4, the n largest gray values from the set of FIG. 4 are found and the corresponding output pels in the 3 x 3 half block are made black or printed on the output image as shown in FIG.5.
Upon the application of both correction techniques de-scribed above an halftone output image may be obtained from a continous tone input image with clearly defined edges on the objects therein and without artifacts ;such as worming or pebbling in the other areas.
- , : :
T he system herein~ desc;ri~b~ed will produce ~printed~ images the same size as the scanned images on printers having three times the linear resolution of the scanner. For ')O YO985-01 2 :
:
example, a scanned image of 200 pels/inch may be printed out on a 600 pels/inch printer. A preferred printer for implementing the invention is the IBM 4250 which has been found to produce excellent electroerosion reproductions suitable for use in the preparation of long life lithographic plates. Various image scanners and other conventional means may be used in the process of capturing the images in a general purpose computer such as an IBM System 370 and the described algorithms 11) may be run by the computer. When it is necessary to change the output image size, the input image may be modified in size by any of a variety of algorithms suit-able for handling digital muiti-level images that are known in the art.
; .
1~ :
~:
: :
: : `
Claims (5)
1. A method for producing a halftone image from a continuous tone input image, the pixels of which have been digitized as gray scale values, using a printer having the capability to print a set of output pel patterns producing a respective set of discrete gray scale values, comprising the steps of:
selecting for printing blocks of pel patterns with gray scale values approximating the gray scale values of the input pixels;
determining the error in a first block due to the difference between the gray scale value associated with the block and the gray scale value of the corresponding input pixel;
distributing said error in said first block to at least two adjacent blocks;
determining the error in one of said adjacent blocks due to the difference between the gray scale value associated with the block and a gray scale value calculated as the sum of the error previously dis-tributed to that block and the gray scale value of the corresponding input pixel; and distributing said error in said one adjacent block to at least two blocks adjacent to it not including said first block; wherein the improvement com-prises:
using a random number generator to determine the distribution of the error between said adjacent blocks such that a random fraction of said error is distributed to one block and the remainder is distributed to the other block.
selecting for printing blocks of pel patterns with gray scale values approximating the gray scale values of the input pixels;
determining the error in a first block due to the difference between the gray scale value associated with the block and the gray scale value of the corresponding input pixel;
distributing said error in said first block to at least two adjacent blocks;
determining the error in one of said adjacent blocks due to the difference between the gray scale value associated with the block and a gray scale value calculated as the sum of the error previously dis-tributed to that block and the gray scale value of the corresponding input pixel; and distributing said error in said one adjacent block to at least two blocks adjacent to it not including said first block; wherein the improvement com-prises:
using a random number generator to determine the distribution of the error between said adjacent blocks such that a random fraction of said error is distributed to one block and the remainder is distributed to the other block.
2. A method for producing a halftone image from a continuous tone input image, the pixels of which have been digitized as gray scale values, using a printer having the capability to print a set of output pel patterns producing a respective set of discrete gray scale values, comprising the steps of:
selecting for printing blocks of pel patterns with gray scale values approximating the gray scale values of the input pixels;
determining the brightness gradient surrounding each input pixel;
setting a threshold value below which said gradient is classified as small and above which it is classi-fied as large;
when said gradient is small, printing the selected discrete block pattern for said pixel; and when said gradient is large, altering the pel pat-tern and printing the block representing said pixel biasing the pels therein toward the darker input pixels.
selecting for printing blocks of pel patterns with gray scale values approximating the gray scale values of the input pixels;
determining the brightness gradient surrounding each input pixel;
setting a threshold value below which said gradient is classified as small and above which it is classi-fied as large;
when said gradient is small, printing the selected discrete block pattern for said pixel; and when said gradient is large, altering the pel pat-tern and printing the block representing said pixel biasing the pels therein toward the darker input pixels.
3. A method for producing a halftone image from a continuous tone input image, the pixels of which have been digitized as gray scale values, using a printer having the capability to print a set of output pel patterns producing a respective set of discrete gray scale values, comprising the steps of:
selecting for printing blocks of pel patterns with gray scale values approximating the gray scale values of the input pixels, said blocks comprising n x n pels:
determining the error in each block due to the difference between the gray scale value associated with the block and the gray scale value of the corresponding input pixel;
distributing said error in each block to at least two adjacent blocks;
using a random number generator to determine the distribution of the error between said adjacent blocks such that a random fraction of said error is distributed to one block and the remainder is distributed to the other block;
determining the brightness gradient surrounding each input pixel;
setting a threshold value below which said gradient is classified as small and above which it is classi-fied as large;
when said gradient is small, printing the n x n block for said pixel; and when said gradient is large, altering the pel pat-tern and printing the n x n block representing said pixel biasing the pels therein toward the darker input pixels.
selecting for printing blocks of pel patterns with gray scale values approximating the gray scale values of the input pixels, said blocks comprising n x n pels:
determining the error in each block due to the difference between the gray scale value associated with the block and the gray scale value of the corresponding input pixel;
distributing said error in each block to at least two adjacent blocks;
using a random number generator to determine the distribution of the error between said adjacent blocks such that a random fraction of said error is distributed to one block and the remainder is distributed to the other block;
determining the brightness gradient surrounding each input pixel;
setting a threshold value below which said gradient is classified as small and above which it is classi-fied as large;
when said gradient is small, printing the n x n block for said pixel; and when said gradient is large, altering the pel pat-tern and printing the n x n block representing said pixel biasing the pels therein toward the darker input pixels.
4. A system for producing a halftone image from a continuous tone input image, the pixels of which have been digitized as gray scale values, using a printer having the capability to print a set of output pel patterns producing a respective set of discrete gray scale values, comprising:
means for selecting for printing blocks of pel pat-terns with gray scale values approximating the gray scale values of the input pixels;
means for determining the error in a first block due to the difference between the gray scale value associated with the block and the gray scale value of the corresponding input pixel;
means for distributing said error in said first block to at least two adjacent blocks;
means for determining the error in one of said adjacent blocks due to the difference between the gray scale value associated with the block and a gray scale value calculated as the sum of the error previously distributed to that block and the gray scale value of the corresponding pixel;
means for distributing said error in said one ad-jacent block to at least two blocks adjacent to it not including said first block; and random number generator means for determining the distribution of the error between said adjacent blocks such that a random fraction of said error is distributed to one block and the remainder is distributed to the other block.
means for selecting for printing blocks of pel pat-terns with gray scale values approximating the gray scale values of the input pixels;
means for determining the error in a first block due to the difference between the gray scale value associated with the block and the gray scale value of the corresponding input pixel;
means for distributing said error in said first block to at least two adjacent blocks;
means for determining the error in one of said adjacent blocks due to the difference between the gray scale value associated with the block and a gray scale value calculated as the sum of the error previously distributed to that block and the gray scale value of the corresponding pixel;
means for distributing said error in said one ad-jacent block to at least two blocks adjacent to it not including said first block; and random number generator means for determining the distribution of the error between said adjacent blocks such that a random fraction of said error is distributed to one block and the remainder is distributed to the other block.
5. A system for producing a halftone image from a continuous tone input image, the pixels of which have been digitized as gray scale values, using a printer having the capability to print a set of output pel patterns producing a respective set of discrete gray scale values, comprising the steps of:
means for selecting for printing blocks of pel pat-terns with gray scale values approximating the gray scale values of the input pixels;
means for determining the brightness gradient surrounding each input pixel;
means for setting a threshold value below which said gradient is classified as small and above which it is classified as large;
means for printing the selected discrete block pattern for said pixel, when said gradient is small;
and means for altering the pel pattern and printing the block representing said pixel biasing the pels therein toward the darker input pixels, when said gradient is large.
means for selecting for printing blocks of pel pat-terns with gray scale values approximating the gray scale values of the input pixels;
means for determining the brightness gradient surrounding each input pixel;
means for setting a threshold value below which said gradient is classified as small and above which it is classified as large;
means for printing the selected discrete block pattern for said pixel, when said gradient is small;
and means for altering the pel pattern and printing the block representing said pixel biasing the pels therein toward the darker input pixels, when said gradient is large.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US06/722,898 US4654721A (en) | 1985-04-12 | 1985-04-12 | System for reproducing multi-level digital images on a bi-level printer of fixed dot size |
| US722,898 | 1985-04-12 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA1248878A true CA1248878A (en) | 1989-01-17 |
Family
ID=24903880
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA000502705A Expired CA1248878A (en) | 1985-04-12 | 1986-02-25 | System for reproducing multi-level digital images on a bi-level printer of fixed dot size |
Country Status (5)
| Country | Link |
|---|---|
| US (1) | US4654721A (en) |
| EP (1) | EP0201674B2 (en) |
| JP (1) | JPS61237574A (en) |
| CA (1) | CA1248878A (en) |
| DE (1) | DE3674943D1 (en) |
Families Citing this family (100)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4742400A (en) * | 1986-02-12 | 1988-05-03 | Ricoh Corporation | Digital image processing device for intermediate tone |
| US4806741A (en) * | 1986-07-16 | 1989-02-21 | Telesis Controls Corporation | Electronic code enhancement for code readers |
| JPS6346863A (en) * | 1986-08-15 | 1988-02-27 | Kenwood Corp | System for receiving picture signal from meteorogical satellite |
| US4680645A (en) * | 1986-08-25 | 1987-07-14 | Hewlett-Packard Company | Method for rendering gray scale images with variable dot sizes |
| US4809082A (en) * | 1986-08-25 | 1989-02-28 | Dainippon Screen Mfg. Co., Ltd. | Method of and apparatus for generating image data used to record an image having gradation |
| US4890167A (en) * | 1986-10-17 | 1989-12-26 | Matsushita Electric Industrial Co., Ltd. | Apparatus for processing image signal |
| US5175807A (en) * | 1986-12-04 | 1992-12-29 | Quantel Limited | Video signal processing with added probabilistic dither |
| DE3785558T3 (en) * | 1986-12-19 | 1997-04-10 | Matsushita Electric Ind Co Ltd | Device for processing signals for displaying images with two levels. |
| US4955065A (en) * | 1987-03-17 | 1990-09-04 | Digital Equipment Corporation | System for producing dithered images from continuous-tone image data |
| JPS63244971A (en) * | 1987-03-30 | 1988-10-12 | Canon Inc | Picture processing method |
| US5164825A (en) * | 1987-03-30 | 1992-11-17 | Canon Kabushiki Kaisha | Image processing method and apparatus for mosaic or similar processing therefor |
| DE3816780A1 (en) * | 1987-05-18 | 1988-12-01 | Canon Kk | Method and device for image processing |
| DE3838730C2 (en) | 1987-11-16 | 1994-07-28 | Canon Kk | Image processing method and apparatus |
| JP2662402B2 (en) * | 1987-11-16 | 1997-10-15 | キヤノン株式会社 | Image processing device |
| US4958218A (en) * | 1987-12-16 | 1990-09-18 | Canon Kabushiki Kaisha | Image processing method and apparatus with dot-processing |
| US4975786A (en) * | 1987-12-28 | 1990-12-04 | Canon Kabushiki Kaisha | Image processing method and apparatus with error diffusion capability |
| JP3077980B2 (en) * | 1988-03-10 | 2000-08-21 | キヤノン株式会社 | Image processing device |
| US4969052A (en) * | 1988-05-11 | 1990-11-06 | Canon Kabushiki Kaisha | Image processing method and apparatus |
| JPH0260771A (en) * | 1988-08-29 | 1990-03-01 | Canon Inc | Image processing system |
| US5086484A (en) * | 1988-08-24 | 1992-02-04 | Canon Kabushiki Kaisha | Image processing apparatus with fixed or variable threshold |
| JP2755307B2 (en) * | 1988-08-29 | 1998-05-20 | キヤノン株式会社 | Image processing device |
| US5153749A (en) * | 1988-09-01 | 1992-10-06 | Canon Kabushiki Kaisha | Image encoding apparatus |
| CA2000705C (en) * | 1988-10-14 | 1999-09-14 | John S. Bowers | Digital half-toning process with error diffusion |
| US5107346A (en) * | 1988-10-14 | 1992-04-21 | Bowers Imaging Technologies, Inc. | Process for providing digital halftone images with random error diffusion |
| US4920501A (en) * | 1988-10-19 | 1990-04-24 | Eastman Kodak Company | Digital halftoning with minimum visual modulation patterns |
| JPH02192966A (en) * | 1989-01-23 | 1990-07-30 | Canon Inc | Image forming device |
| JP2793205B2 (en) * | 1988-10-20 | 1998-09-03 | キヤノン株式会社 | Image processing method |
| US5162925A (en) * | 1988-11-17 | 1992-11-10 | Canon Kabushiki Kaisha | Color image processor capable of performing masking using a reduced number of bits |
| NL8902654A (en) * | 1988-11-24 | 1990-06-18 | Oce Nederland Bv | METHOD AND APPARATUS FOR PROCESSING AND DISPLAYING IMAGE INFORMATION. |
| US5051844A (en) * | 1989-01-30 | 1991-09-24 | Eastman Kodak Company | Digital halftoning with error diffusion |
| US5185602A (en) * | 1989-04-10 | 1993-02-09 | Cirrus Logic, Inc. | Method and apparatus for producing perception of high quality grayscale shading on digitally commanded displays |
| US5201013A (en) * | 1989-04-24 | 1993-04-06 | Ezel, Inc. | Dither processing method |
| US4926267A (en) * | 1989-05-26 | 1990-05-15 | Nynex Corporation | Reproduction of halftone original with reduced moire |
| US5053887A (en) * | 1989-06-27 | 1991-10-01 | Ibm Corporation | Technique for producing a fine grained dithered halftone image having an increased number of gray levels |
| JPH0380767A (en) * | 1989-08-24 | 1991-04-05 | Ricoh Co Ltd | Gradation recorder for image |
| US5070413A (en) * | 1989-10-10 | 1991-12-03 | Eastman Kodak Company | Color digital halftoning with vector error diffusion |
| US5051841A (en) * | 1989-10-16 | 1991-09-24 | Bowers Imaging Technologies, Inc. | Process for providing digital halftone images with random error diffusion |
| US5087981A (en) * | 1990-01-02 | 1992-02-11 | Eastman Kodak Company | Error diffusion of overlapping dots |
| US4959730A (en) * | 1990-01-02 | 1990-09-25 | Eastman Kodak Company | False density contour suppression using randomly modified input signals for comparison to threshold values |
| US5055942A (en) * | 1990-02-06 | 1991-10-08 | Levien Raphael L | Photographic image reproduction device using digital halftoning to screen images allowing adjustable coarseness |
| US5214517A (en) * | 1990-02-07 | 1993-05-25 | Eastman Kodak Company | Digital halftoning with correlated minimum visual modulation patterns |
| US5031050A (en) * | 1990-02-26 | 1991-07-09 | Hewlett-Packard Company | Method and system for reproducing monochromatic and color images using ordered dither and error diffusion |
| US5260806A (en) * | 1990-08-29 | 1993-11-09 | E. I. Du Pont De Nemours And Company | Process for controlling tone reproduction |
| US5140432A (en) * | 1990-09-12 | 1992-08-18 | Hewlett-Packard Company | Method and system for printing in one or more color planes with improved control of error diffusion |
| US5226094A (en) * | 1990-10-19 | 1993-07-06 | Xerox Corporation | Method for making image conversions with error diffusion |
| GB2249910B (en) * | 1990-11-19 | 1994-11-30 | Le Proizv Ob Poligraficheskogo | Method of and apparatus for making a half-tone reproduction of an image |
| US5341228A (en) * | 1990-12-04 | 1994-08-23 | Research Corporation Technologies | Method and apparatus for halftone rendering of a gray scale image using a blue noise mask |
| JPH04304073A (en) * | 1991-03-30 | 1992-10-27 | Toshiba Corp | Picture processor |
| US5130823A (en) * | 1991-06-27 | 1992-07-14 | Cactus | Error diffusion system |
| US5276535A (en) * | 1991-12-24 | 1994-01-04 | Levien Raphael L | Method and apparatus for halftoning of images using grayscale error diffusion |
| US5226096A (en) * | 1991-10-11 | 1993-07-06 | Xerox Corporation | Digital halftoning with selectively applied dot-to-dot error diffusion |
| US5268774A (en) * | 1991-11-27 | 1993-12-07 | Xerox Corporation | Halftoning with enhanced dynamic range and edge enhanced error diffusion |
| US5243443A (en) * | 1991-12-06 | 1993-09-07 | Xerox Corporation | Halftoning with error feedback and image dependent enhancement |
| US5245678A (en) * | 1991-12-20 | 1993-09-14 | Xerox Corporation | Image conversion with lossy adaptive error diffusion |
| US5751265A (en) * | 1991-12-24 | 1998-05-12 | Cirrus Logic, Inc. | Apparatus and method for producing shaded images on display screens |
| WO1993013513A1 (en) * | 1991-12-24 | 1993-07-08 | Cirrus Logic, Inc. | Process for producing shaded images on display screens |
| US5331429A (en) * | 1992-02-07 | 1994-07-19 | Levien Raphael L | Digital generation of halftone images with error diffusion and frequency matched periodic screen rulings |
| AU656057B2 (en) * | 1992-03-27 | 1995-01-19 | Milliken & Company | Method and apparatus for reproducing blended colorants on an electronic display |
| US5260807A (en) * | 1992-06-05 | 1993-11-09 | Eastman Kodak Company | Method and apparatus for imbedding controlled structure for gray scale rendering |
| US5363213A (en) * | 1992-06-08 | 1994-11-08 | Xerox Corporation | Unquantized resolution conversion of bitmap images using error diffusion |
| US5337160A (en) * | 1992-07-01 | 1994-08-09 | Hewlett-Packard | Error diffusion processor and method for converting a grey scale pixel image to a binary value pixel image |
| US5374997A (en) * | 1992-07-31 | 1994-12-20 | Xerox Corporation | High addressability error diffusion with minimum mark size |
| US5539667A (en) * | 1992-09-15 | 1996-07-23 | Gcc Technologies | Method and apparatus for improved digital film recorder |
| US5489991A (en) * | 1992-10-09 | 1996-02-06 | International Business Machines Corporation | Method for generating digital halftone images |
| US5321525A (en) * | 1992-12-14 | 1994-06-14 | Xerox Corporation | Clustered halftoning with dot-to-dot error diffusion |
| US5325211A (en) * | 1993-01-04 | 1994-06-28 | Xerox Corporation | Error diffusion with output and input based feedback |
| EP0612181A1 (en) * | 1993-02-19 | 1994-08-24 | Linotype-Hell Ag | Method and appartus for converting grey scale values into binary values |
| DE59309785D1 (en) * | 1993-04-08 | 1999-10-21 | Heidelberger Druckmasch Ag | Process for the optimization of frequency-modulated grids using threshold mountains |
| NL9300684A (en) * | 1993-04-22 | 1994-11-16 | Oce Nederland Bv | Method of halftoning digitized gray value images and image processing apparatus suitable for performing such a method. |
| JP3603906B2 (en) * | 1994-04-15 | 2004-12-22 | 富士写真フイルム株式会社 | Image signal binarization processing apparatus and method |
| US5592592A (en) * | 1994-07-01 | 1997-01-07 | Seiko Epson Corporation | Method and apparatus for minimizing artifacts in images produced by error diffusion halftoning utilizing ink reduction processing |
| US5692109A (en) * | 1994-07-01 | 1997-11-25 | Seiko Epson Corporation | Method and apparatus for minimizing artifacts in images produced by error diffusion halftoning |
| JPH08204971A (en) * | 1994-10-31 | 1996-08-09 | Xerox Corp | Image compression method using predictive coding and error diffusion |
| US5493416A (en) * | 1994-10-31 | 1996-02-20 | Xerox Corporation | Method combining error diffusion and traditional halftoning with arbitrary screen orientation |
| EP0712235B1 (en) * | 1994-11-08 | 2002-05-02 | Xerox Corporation | Pixel pair grid halftoning for a hyperacuity printer |
| US5581371A (en) * | 1995-03-07 | 1996-12-03 | Eastman Kodak Company | Error diffusion method |
| US5757517A (en) * | 1995-03-23 | 1998-05-26 | Eastman Kodak Company | Adaptive error diffusion method |
| JP3396992B2 (en) * | 1995-03-30 | 2003-04-14 | セイコーエプソン株式会社 | Image processing method and apparatus |
| NL1001788C2 (en) * | 1995-11-30 | 1997-06-04 | Oce Tech Bv | Method and image reproduction device for reproducing grayscale. |
| US5809177A (en) * | 1996-06-06 | 1998-09-15 | Xerox Corporation | Hybrid error diffusion pattern shifting reduction using programmable threshold perturbation |
| US5867632A (en) * | 1996-11-07 | 1999-02-02 | Xionics Document Technologies, Inc. | System and method for providing glossy compression and contone/bitonal image conversion |
| US6211859B1 (en) | 1997-03-10 | 2001-04-03 | Chips & Technologies, Llc | Method for reducing pulsing on liquid crystal displays |
| US6034663A (en) * | 1997-03-10 | 2000-03-07 | Chips & Technologies, Llc | Method for providing grey scale images to the visible limit on liquid crystal displays |
| US6449061B2 (en) * | 1997-09-23 | 2002-09-10 | Xerox Corporation | System and method for providing dynamic noise profile selection for hybrid and error diffusion image processing |
| US6181357B1 (en) | 1998-02-17 | 2001-01-30 | International Business Machines Corporation | Method and system for the modification of the timing of a plurality of PEL clocks in an electrophotographic device |
| US6753978B1 (en) | 1999-09-22 | 2004-06-22 | Sharp Laboratories Of America, Inc. | Higher order error diffusion of digital halftoning |
| JP2001157056A (en) * | 1999-11-30 | 2001-06-08 | Fujitsu Ltd | Method and device for producing halftone, and computer- readable recording medium with halftone making program thereon |
| US6707576B1 (en) | 1999-12-02 | 2004-03-16 | Sharp Laboratories Of America, Inc. | Noise modulation error diffusion of digital halftoning |
| JP3912055B2 (en) * | 2000-10-06 | 2007-05-09 | セイコーエプソン株式会社 | Image processing apparatus, print control apparatus, image processing method, and recording medium |
| JP3666427B2 (en) * | 2000-10-06 | 2005-06-29 | セイコーエプソン株式会社 | Image processing apparatus, print control apparatus, image processing method, and recording medium |
| DE50103922D1 (en) * | 2000-12-08 | 2004-11-04 | Oce Printing Systems Gmbh | METHOD FOR HALF-TONE REPRESENTATION OF AN IMAGE, AND IMAGE PROCESSING DEVICE AND PRINTING DEVICE FOR CARRYING OUT THIS METHOD |
| CN1213592C (en) * | 2001-07-31 | 2005-08-03 | 佳能株式会社 | Adaptive two-valued image processing method and equipment |
| US7268919B2 (en) * | 2002-01-17 | 2007-09-11 | Seiko Epson Corporation | Image data processing apparatus, method, and program that diffuses gradiation error for each pixel in target block |
| US7265872B2 (en) * | 2003-10-15 | 2007-09-04 | International Business Machines Corporation | Error diffusion with averaged directional biases |
| US7826682B2 (en) * | 2005-04-14 | 2010-11-02 | Agfa Healthcare | Method of suppressing a periodical pattern in an image |
| US7675648B2 (en) * | 2006-06-22 | 2010-03-09 | Infoprint Solutions Company, Llc | Method and apparatus for converting a color representation of an image to a grayscale representation |
| GB2473476B (en) * | 2009-09-11 | 2015-01-07 | Andrew William Peter Cave | Computer program for generating 1-bit image data from multiple-bit image data |
| GB2473475B (en) | 2009-09-11 | 2014-12-17 | Andrew William Peter Cave | Computer program for generating 1-bit image data from multiple-bit image data |
| JP5482535B2 (en) * | 2010-07-23 | 2014-05-07 | コニカミノルタ株式会社 | Image processing apparatus and image processing method |
| WO2020112099A1 (en) * | 2018-11-28 | 2020-06-04 | Hewlett-Packard Development Company, L.P. | Printing liquid colors |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4084183A (en) * | 1970-03-18 | 1978-04-11 | Dr. Ing. Rudolf Hell Gmbh. | Method for the electro-optical reproduction of half-tone pictures |
| US4110795A (en) * | 1976-12-20 | 1978-08-29 | Litton Systems, Inc. | Method of graphic data redundancy reduction in an optical facsimile system |
| JPS55137530A (en) * | 1979-04-13 | 1980-10-27 | Dainippon Screen Mfg Co Ltd | Color separation plate recording method for plural original pictures by image signal memory system color scanner |
| DE3003607C2 (en) * | 1980-02-01 | 1984-01-05 | Dr.-Ing. Rudolf Hell Gmbh, 2300 Kiel | Circuit arrangement for partial post-correction of color recognition spaces during color recognition |
| JPS57123766A (en) * | 1981-01-23 | 1982-08-02 | Ricoh Co Ltd | Density gradation processing system of digital facsimile |
| US4495522A (en) * | 1981-07-07 | 1985-01-22 | Konishiroku Photo Industry Co., Ltd. | Recording apparatus and method of picture image |
| JPS5810960A (en) * | 1981-07-11 | 1983-01-21 | Ricoh Co Ltd | Picture signal processing circuit |
| US4430668A (en) * | 1981-08-10 | 1984-02-07 | Miles Jr Ernest P | Digital FACSIMILES (Fast Additive Color Separation Internegatives MILES) |
| US4463374A (en) * | 1981-11-20 | 1984-07-31 | Trilog, Inc. | Color plotter interface and control circuit |
| JPS59216371A (en) * | 1983-05-24 | 1984-12-06 | Nippon Telegr & Teleph Corp <Ntt> | Picture expressing method |
| JPS60153264A (en) * | 1984-01-20 | 1985-08-12 | Ricoh Co Ltd | Transmission system of half tone picture |
-
1985
- 1985-04-12 US US06/722,898 patent/US4654721A/en not_active Expired - Fee Related
-
1986
- 1986-02-25 CA CA000502705A patent/CA1248878A/en not_active Expired
- 1986-02-28 EP EP86102646A patent/EP0201674B2/en not_active Expired - Lifetime
- 1986-02-28 DE DE8686102646T patent/DE3674943D1/en not_active Expired - Fee Related
- 1986-03-28 JP JP61068890A patent/JPS61237574A/en active Granted
Also Published As
| Publication number | Publication date |
|---|---|
| US4654721A (en) | 1987-03-31 |
| EP0201674A3 (en) | 1987-09-16 |
| DE3674943D1 (en) | 1990-11-22 |
| EP0201674A2 (en) | 1986-11-20 |
| EP0201674B1 (en) | 1990-10-17 |
| JPS61237574A (en) | 1986-10-22 |
| JPH0472433B2 (en) | 1992-11-18 |
| EP0201674B2 (en) | 1994-04-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA1248878A (en) | System for reproducing multi-level digital images on a bi-level printer of fixed dot size | |
| EP0749232B1 (en) | Dither matrix for halftone rendering of a gray scale image using a blue noise mask | |
| US5341228A (en) | Method and apparatus for halftone rendering of a gray scale image using a blue noise mask | |
| US5323245A (en) | Perpendicular, unequal frequency non-conventional screen patterns for electronic halftone generation | |
| US5583660A (en) | Non-perpendicular, equal frequency non-conventional screen patterns for electronic halftone generation | |
| EP0481602B1 (en) | A general kernel function for electronic halftone generation | |
| JPH0774950A (en) | Generating method for half-tone picture using modeling of print symbol | |
| US5489991A (en) | Method for generating digital halftone images | |
| JPH0336876A (en) | Half-toned pell pattern producing method | |
| Goertzel et al. | Digital halftoning on the IBM 4250 printer | |
| EP0254448B1 (en) | Optimal color half-tone patterns for raster-scan images | |
| US5051841A (en) | Process for providing digital halftone images with random error diffusion | |
| JPH06105129A (en) | Apparatus and method for electronic hi-fi screenless conversion using separable filter | |
| US5264926A (en) | Perpendicular, equal frequency non-conventional screen patterns for electronic halftone generation | |
| US6002493A (en) | Method for reproducing an image | |
| US5258832A (en) | Non-perpendicular, unequal frequency non-conventional screen patterns for electronic halftone generation | |
| EP0361538A1 (en) | Method and system for edge enhancement in reproducing multi-level digital images on a bi-level printer of fixed dot size | |
| CA2000705C (en) | Digital half-toning process with error diffusion | |
| Wu et al. | Supercell dither masks with constrained blue noise interpolation | |
| US5764810A (en) | Screenless conversion of continuous tone images with alterable dot spacing patterns | |
| CA2231820C (en) | Method and apparatus for halftone rendering of a gray scale image using a blue noise mask |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MKEX | Expiry |