CA1047935A - Pseudo halftone print generator and method - Google Patents
Pseudo halftone print generator and methodInfo
- Publication number
- CA1047935A CA1047935A CA241,587A CA241587A CA1047935A CA 1047935 A CA1047935 A CA 1047935A CA 241587 A CA241587 A CA 241587A CA 1047935 A CA1047935 A CA 1047935A
- Authority
- CA
- Canada
- Prior art keywords
- gray scale
- count
- scale levels
- supplying
- 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
Abstract
PSEUDO HALFTONE PRINT GENERATOR AND METHOD
Abstract Gray scale levels are converted to pseudo halftone print by a pseudo halftone generator which reduces Moire effects. The generator includes a comparator and an incrementing cyclic count-ing mechanism to compare the gray scale level to the cyclicly generated reference count for each print position in a print line to generate a print signal upon the level being greater than the count. The cyclicly generated count is incremented by a preselected number N for each print position and recycles at the square of that number (N2). The square value may also be selected as the maximum level of the gray scale range. N adjacent print lines may be grouped together to form a series of (NxN) matrices. The refer-ence or lowest count value for each of the lines in a matrix is different and they range from zero to N-1. The arrangement of the lines in the matrix is such as to minimize the difference be-tween the sums of the count element values for any two adjacent print lines. The count for each generator is initialized at the beginning of the print line and may be established randomly for each print line or, alternatively, for each matrix of print lines, the matrix being arranged in a predetermined manner.
Abstract Gray scale levels are converted to pseudo halftone print by a pseudo halftone generator which reduces Moire effects. The generator includes a comparator and an incrementing cyclic count-ing mechanism to compare the gray scale level to the cyclicly generated reference count for each print position in a print line to generate a print signal upon the level being greater than the count. The cyclicly generated count is incremented by a preselected number N for each print position and recycles at the square of that number (N2). The square value may also be selected as the maximum level of the gray scale range. N adjacent print lines may be grouped together to form a series of (NxN) matrices. The refer-ence or lowest count value for each of the lines in a matrix is different and they range from zero to N-1. The arrangement of the lines in the matrix is such as to minimize the difference be-tween the sums of the count element values for any two adjacent print lines. The count for each generator is initialized at the beginning of the print line and may be established randomly for each print line or, alternatively, for each matrix of print lines, the matrix being arranged in a predetermined manner.
Description
'7935 1 Back~round of th~ Invention In the graphic arts industry9 which ranges from the publication of newspapers, books, and magazines to computer output printers, the printing process commonly used is an all or nothing binary process. This all or nothing process comprises depositing a dot of ink on paper wherever it is desired to print and to deposit no ink where the absence of an image is des;red. This all or nothing process ;s acceptable and in fact desirable when alphanumeric characters or other symbols are printed. A problems arises, however, when images employing gray scales or light gradations in continusus tones are to be printed, such as in the printing of photographs. This prob-lem is commonly solved by transforming the continuous tone of the original image into half~one or pseudo halftone images. ~lalftone images typically comprise a large number of ink dots of various sizes. The size of the ink dots correspond to the shades or tones to be reproduced. When the dots in the corresponding spaces on the paper between the dots are small compared to the visual acuity of the human eye, they are subliminal to the eye and not recognized. The dots and $he spaces on the paper thus fuse visually and ;
trick the eye into believing that various shades of continuous tones are seen. Pseudo halftone images refers to the process of reproducing the con-tinuous tone images or gray scale with a printing device having a fixed ~ `
printed spot size and fixed spot-to-sPt spacing- The level of gray to be ~i reproduced thus is represented by a number of dots which are printed out of a specified line segment or matrix array of printing positions. If the individual spots are sufficiently small, they effect a merger insofar as the eye is concerned to ~orm a visual merger with the spaces between the dots to cause the eye to believe it is seeing various shades of continuous tones.
Electronics have been applied to image printing at various times with respect to the creation of half tone or pseudo half tone images. One example is Richard G. Sweet et al U.S. patent No. 3,373,437, issued May 14 1968 entitled "Fluid Droplet Recorder with a Plurality of Jets." This patent illustrates an ink jet printing mechanism wherein jets of ink are each caused to break into a uniform stream of droplets which are directed to , ~ , ~ 7935 1 impact the paper which ;s to be printed. Selected droplets wh;ch are not to be printed are charged as they break from the stream and are deflected by a constant electrostatic field to impact a gutter. A video intelligence sig-nal representing the relative brightness or inverse gray scale of the image to be reproduced ;s sampled and applied to the charging electrode into a R-C time constant circuit. The sampled pulse thus decays at a rate deter-mined by the R-C time constant. So long as the charging electrode stays above a predetermined threshold value, the droplets emanating fro~ the jet are charged and deflected to the gutter, leaving the paper unmarked. Upon the sig- -nal dropping below the threshold value, the rema;ning drops of the stream are uncharged and impact the paper. The ratio of drops impacting the paper with respect to the total number of drops produced prior to application of the next ~ -video signal comprises the gray scale level of the sampled video signal. In a printing line made up of a string of ink droplets or their unprinted posi- -tions, the droplets would thus be grouped periodically in accordance with the periods of the video signal, the droplets appearing towards the end of each period and the spaces at the beginning of each period.
Another example is shown in IBM Technical Disclosure Bulletin, "Pseudo Half Tone for Representing Continuous Tone Images in Black-While Facsimile ; -Systems," G~K. Machol, Vol. 9, No. 6, Nov. 1966, pages 636-637. This example illustrates the derivation of pseudo half tone matrices in accordance with a -~
predetermined logic. The analog Yideo si3nal is applied to a quantizer which samples the ~ideo signal and provides a binary coded output indicating the ~ ~ -gray scale level of the sample point. The logic circluit then transforms the binary coded values into a white zero or black one for each print position of `
the matrix. Also shown is the incorporation of an alternating bit in the logic to allow estimation of an eight-level quantization by a four bit matrix through the incorporation of an extra black dot in every other matrix. In U.S. Patent No. 3,604,846, issued Dec. 26, 1972, David Behane et al, entitled "Method and System for Reconstruction of Half-Tone Images," the matrix pat-terns are placed in a table in storage and are accessed in accordance with a quantized video density value.
. . . .~ , , . , ~
~4~47935 1 The difficulty with the dot bunching arrangement of Sweet et al is that it generates pronounced Moire patterns in any larger area of the same or similar gray scale levels. Judicious choice of the matrices of Machol or of Behane et al may reduce but do not eliminate ~he Moire patterns. They, how-ever, introduce the requirement for complex logic circuitry or for an exten-sive storage and accessing mechanism to transform the quantized gray level into the proper form.
5ummary of the Invention It is therefore an object of the present invention to employ a simpli-fied arrangement to generate pseudo halftone images which reduces Moire pat-tern effects.
Another object of the present ~nvention is to provide an even distribu-tion of spots for a glven gray scale leYel such that the resultant image may ~ -be scanned and reproduced without significant variation.
According to the present invention, a pseudo halftone generator is initiated at the beginning of each line of printing in accordance with a ran~
dom number generator. The pseudo halftone generator is arranged to count cyclicly from the initiated count, incrementing the count by a preselected number N for each print position and recycling at the square o~ that number.
For each prin~:position, the count is compared to the gray scale level input to generate a print signal upon the level being greater than the count. Fur-ther in accordance with the present invention, an adjacent group of print lines may be arranged in a preselected matrix format of dimension N controlled by the initiation so that the lowest or reference count value for each line in the matrix is different and is shifted from the lowest count value of the pre-ceding line in the matrix by the number of print positions equal to an inte-ger close to the square root of N.
A feature of the present invention is that it is operable with the presentation of gray scale levels at a different frequency than the recycling frequency of the pseudo halftone generator.
The foregoing and other objects, features and advantages of the inven-tion will be apparent from the following more particular description of the ~L04793~;
1 preferred embodlment of t'ne invention as illustra~ed ~n the accompanying draw-ings.
Brief Description of the Drawings FIGURE 1 is a block diagram of a pseudo halftone generator constructed in accordance with the invention.
FIGURE 2 is a diagrammatic represen~ation of gray scale matrices.
FIGURE 3 is a diagrammatic representation of the matrices of Figure with the appropriate dot patterns.
FIGURE 4 is a schematic block diagram of a scanning and printing system employing the pseudo halftone generator of Figure 1.
FIGURE 5 is a block diagram of alternative initiation circuitry for the system of Figure 4.
FIGURE 6 is a diagrammatic representation of a series of gray scale matrjces.
FIGURE 7 is a diagrammatic representation of another series of gray scale matrices. ~-FIGURE 8 is a diagrammatic representation of a gray scale matrix and its corresponding weight distribution matrix.
FIGURE 9 is a diagrammatic representation of a gray scale matrix and its corresponding dot pattern.
FIGURE 10 is a diagrammatic representation of an exemplary gray scale matrix showing the low count value print position in each line.
FIGURE 11 is a diagrammatic representation of a series of various size gray scale matrices and showing the low count value print position for each line in accordance with the invention.
FIGURE 12 is a schematic block diagram o~ a series of pseudo halftone generators arranged in parallel in accordance with ~he invention.
Detailed Description of the Invention Pseudo halftone as discussed herein refers to the process of reproducing continuous tone images or gray scales with a printing device having only two levels of gray, namely, black and white, and operating with a fixed printed spot size and fixed spot-to-spot spacing. An embodiment of the pseudo half-, ." -. .- - . . ,; . . ., , , - , - .
- 11)47935 1 tone generator of the present inventlon is illustrated in Figure 1. In the example shown, the quantized, b~nary coded gray scale level is supplied to digital comparator 10 on lines 11. A cyclicly generated binary coded count is supplied to the comparator 10 on lines 12 through 15. Such digital com-parators are well known in the art and may be arranged to provide an output on line 16 when the digital value on lines 11 exceeds that oF lines l2-15.
Thus, whenever the gray scale input level exceeds the count appearing on lines 12-15, comparator 10 supplies an output signal on line 16 to operate the printing mechanism to print a dot in the corresponding print position.
At the beginning of a print line, an initiation count is supplied on line 17.
The initiation count may be achieved by random number generation or in a pre-determined manner, both of which will be discussed hereinafter. The signals on line 17 are applied to flip-flops 18 and 19 and to input lines 14 and 15 of comparator 10. The flip-flops 18 and 19 are selectively set or reset in ac-cordance with the signal appearing at the inputs 20 and 21 thereof. The resul-tant output signals, if any, are supplied on lines 12 and 13 to comparator 10.
For example, a positive signal appearing on any one of the lines may repre-sent the presence of a binary bit, whereas a negative signal on any one of the lines may represent the absence of a binary b;t. For each print position in a printing line of the printing apparatus, a pulse is supplied on line 24.
The pulse causes flip-flop 19 to reverse its condition, for example, changing from the state of supplying a binary 1 on line 13 to the state of supplying a 0 on line 13. When switching from the 1 to the 0 state, flip-flop 19 fur-ther supplies a carry output pulse on llne 25 to flip-flop 18. The occurrence of a pulse on line 25 similarly causes flip-flop 18 to reverse state. Thus, the net effect of a pulse appearing on line 24 is to add the binary equivalent of 4 to the count appearing at the input lines 12-15 to comparator 10. Thus, N is equal to 4. In the example shown, the count appearing on lines 12-15 to comparator 10 may range from a low of 0 to a high value, represented by all ones, of 15. Inasmuch as no carry output is supplied from flip-flop 18 to a higher order input to comparator 10, the effect of a pulse on line 24 when both flip-flops 18 and 19 are in the 1 state is to reverse both flip-flops SA9-74-038 ~6-1 to the 0 state. This has the e-ffect of recycling the count value appearing at input 12-15 upon that value otherwise reaching a value of 16 or greater.
For example, if the coun~ represented by lines 12 ~hrough 15 were a binary 1 (1110), the appearance o~ a pulse on line 24 wlll operate both ~lip-flops 18 and 19 to the opposite state so that the resultant count appearing on lines 12-15 is the binary equiYalent of 2 (0010). The count supplied on lines 12-15 can thus be set to be incremented by a selected number of 4 and to recycle at the count of 16, which value is the square of the preselected incrementing number N.
The exemplary gray scale range represented b~ lines 11 sùpplied to com-parator 10 is from a level of zero (all white) to 16 (all black~. Five lines are therefore required to indicate the highest gray scale level of 16, which takes the form 10000. If 1t is desired to employ only 16 gray scale levels from 0 to 15, a logic cîrcuit may be added to automatically con~ert the 15 level (1111) to the 16 level (10000).
Referring to Figure 2, some print line matrices are illustrated. That of Figure 2A corresponds to a realized pattern produced by the circuitry of Figure 1 wherein each print line is one of the rows of the Figure. With respect to row 30, the circuitry would be initiated with a count of 0 and incremented by four for each subsequent print position, recycling to 0 rather than counting 16. Row 31 would be initiated at a count of 11 and similarly incremented by four for each print;position. Row 32 is seen initiated with a count of 1 and also incremented byv four for each print position. Lastly, row 33 is initiated with a count of 10 and similarly incremented. The number in each print position of the matrices thus indicates the gray scale level which must be exceeded in order for a dot to be printed at that print position.
The pattern illustrated in the matrices of Figure 2B is different and is simply sequential in nature and is not arranged to be produced by circuitry resembling that of Figure 1.
Figure 3 illustrates exemplary dot patterns resulting from the same gray level input as printed in accordance with the matrices shown ln Figure 2. In the dot pattern of Figure 3A, the gray level value of 4 supplied at input S~*74-038 -7-, . . .
,. , ~ -; ~. - , - :
- ~4793S
1 lines 11 to comparator 10 causes the comparator to provide an output at print position 0, but exceeds the count supplied on lines 12-15 in no other position on line 30. Similarly, the gray scale level of 4 exceeds only the count of 3 in line 31, exceeds only the supplied count of 1 in line 32 and exceeds only the count of 2 in line 33. The net effect of the apparatus of Figure 1 is to scatter the dots throughout the matrix and thus avoid bunch-ing and a tendancy to create Moire patterns.
The arrangement of print position gray scale levels shown in Figure 2B
is not in accordance with the present invention and results in dot patterns as shown in Figure 3B. This dot pattern in essence alters the gray level to a combination of a black line and white space. Other gray scale levels applied to the matrix of Figure 2B similarly result in dot patterns having bunching of the dots and in the creation of Moire patterns.
Referring to Figure 4, a system is illustrated which employs the pseudo halftone generator 40 shown in detail in Figure 1. The pseudo halftone gener-ator may be employed to convert gray scale levels for printing either from the output of a scanner 41 as quantized by quantizer 42 or from a computer system 43. A switch 44 may selectively supply the data from the scanner and quantizer or from the computer system to a buffer 45. The switch 44 may alternatively supply the quantized data from quantizer 42 to the computer system 43 for storage in the memory thereof. The computer system may then selectively transmit the quantized gray scale information from the memory to the buffer 45. Buffer 45 is required because the data transmission char-acteristics of the quantizer 42 or o~ the computer system 43 may not be identical to that of the pseudo halftone generator 40. The gray scale data is supplied by 6uffer 45 on lines 11 to the pseudo halftone generator 40. The output of the generator is supplied on line 16 to a printer 46. The printer 46 may comprise an ink jet printer, a cathode ray microfilm printer, a wire pr;nter or even an impact printer or typewriter printer fitted with special spot printing elements. All such printers having a mechanism such as line start indicator 47 to indicate the termination of printing of one line and the beginning of printing of another line. The line start indicator 47 is .
~047935 1 connected to buf~er 45 via line 48 and to a random number generator 50 via line 51. Random number generators normally comprise a recycling counting mechanism such as a binary counter which is stepped or sequenced by accumu-lated noise signals exceeding threshold levels and are unclocked. Thus, the number generated is random as compared to any of the machine cycles or clock-ing of the receiving mechanism. The signal appearing on line 51 from line start indicator operates the random number generator to supply its then ran-domly selected output number on lines 17 to the initiating input of pseudo halftone generator 40. Clock 52 supplies a pulse on line 24 for each print position of printer 46. Synchronism between clock 52 and the printer 46 is obtained by virtue of either feedback from the printer on line 53, or feed-forward ~rom the clock 52 to drive the operation of printer 46 via line 54.
The clock 52 is arranged to provide appropriate signals on line 56 to drive buffer 54 in accordance with the relative periodicity or frequency of gray scale inputs relative to the printing rate of printer 46. For example, the clock signals supplied on line 56 may be one-half or may be one-fourth the frequency of those supplied on line 24 to represent that the gray scale levels change once every two print positions of printer 46. Except for the pseudo halftone generator 40 described in Figure 1, all the elements of Figure 4 are individually well known in the art and need not be described in detail here.
In operation, scanner 41 and quantizer 42 or computer system 43 deliver gray scale data, via switch 44~ to buffer 45. Upon line start indicator 47 of printer 46 indicating the start of a line by means of a signal on line 483 buffer 45 delivers gray scale data on lines 11 under the control of clock signal appearing on line 56. The line start indicator signal appearing on line 51 also operates random number generator 50 to initiate the pseudo half-tone generator by signals on line 17. The pseudo halftone generator 40 responds by comparing the initial count to the gray scale level and supplying ~;
the pseudo halftone output on line 16 in serial form, advancing the count sup-plied to comparator 10 in accordance with the pulses supplied on line 24 from clock 52.
SA9-74-0~8 -9-` iL~4l7 9 3~j 1 The arrangement of Figure 5 comprises an alternative to the randomnumber generator 50 in Figure 4. With reference to Figure 2A, use of the random number generator 50 will result in random starting points for each of the lines 30 through 33 rather than the uniform matrix pattern shown in the figure. This means that each print line will begln with any randomly selected number from 0 to 15.
To implement a pre-formatted matrix, the apparatus of Figure 5 is required. There, the line start indicator signal on line 51 is supplied to the input of a ring or four phase clock circuit 59. Line 60 of ring 59 is connected to random number generator 65~ to gate circuit 66, and to the "load"
inputs of registers 67 through 69. Lines 61 through 63 are connected, res-pectively, to gate circuits 71 through 73. Random number generator 65 is the same as random number generator 50 in Figure 4 except that the lowest order stages of the counter are preset and the noise excitation is applied to the binary 4 (N3 stage. Thus, the random number generator 65 randomly selects from values 0, 4, 8 or 12. The generator 65 is connected via lines 75 to gate circuit 66 and to adders 77 through 79. Gate circuits 66 and 71-73 are connected to lines 17 of Figures ~ and 1.
In operation, assume that ring 59 is in the last state, number 3. A
subsequent line start indicator signal on line 51 operates the ring to the "zero" state, providing a similar pulse signal on line 60. This signal operates the random number generator 65 to provide a randomly selected num-ber on lines 75. As an example, the numbers which may be randomly selected may be 0, 4, 8 or 12 employed in line 30 of Figure 2A as discussed above.
The randomly selected number is supplied to gate circuit 66, and adders 77 through 79 Each adder adds a different predetermined value to the randomly selected number and supplies the resultant output, respectively, to registers 67 through 69. To accomplish the matrix shown in Figure 2A, adder 77 must add the value of 11 to the randomly selected number on line 75 without a carry. This means that any total of binary 16 or greater will appear as a value of 0 or greater. For example, should the selected number on line 75 be 8, the total from adder 77 w;th a carry would be a binary 19, but appears - , ~ ;~ . . -~C~47935 1 as an output of binary 3 without the carry. Similarly, adder 78 adds a Yalue of 1 to the same randomly selected number on line 75 and adder 79 adds a value of 10 to the same randomly selected number. The signal on line 60 operates gate 66 to transmit the randomly selected number to lines 17 to `thereby ;n;tiate the pseudo halftone generator 40. The signal on line 60 also operates the "load" inputs of registers 67 through 69 to load therein the outputs of the respective adders 77-79. At the next line start indicator signal on line 51, the ring advances to the "1" state, providing an output signal on line 61. This signal operates gate circuit 71 to transmit the total appearing in register 67 to lines 17 of the pseudo halftone generator 401 thereby initiating the generator for print line 31 in Figure 2A. The next line start indicator signal on line 51 operates the ring 59 to state "2" to provide an output signal on line 62. This signal operates gate 72 to transmit the total from register 68 to the initiate lines 17 of the half-tone generator 40. This causes the halftone generator 40 to produce the print line 32 of Figure 2A. Las~ly, the next line start indicator signal on ~` !
line 51 operates the ring to condition "3" to provide an output signal on line 63. This signal operates gate circuit 73 to provide the total from register 59 to the initiate input lines 17 of the halftone generator 40. The halftone generator then produces the print line 33 shown in Figure 2A. The next line start indicator signal on line 51 again operates the ring to state "O" to cause the generation of another randomly selected number. As an alter~
native, registers 67 through 69 may be eliminated and line 80 employed to connect line 51 directly to the random number generator 65. This causes selection of one of the random numbers in line 30 of Figure 2A for each line start indicator signal on line 51. Operation with the registers 67 through 69 and without line 80 causes the generation of a regular matrix pattern sëlected to avoid Moire ~atterns as much as possible as shown in Figure 2A.
Operation of the system with line 80 and without registers 67-69 allows con-siderable shifting of the print lines 31 through 33 individually with respect to print line 30. This may result in a more random relative location of the dot pattern such as shown in Figure 3A which, if truly random, results in a ` ,,.. . , - ,, .. ,, , . , ., . . , ~ .. . .
~47g3S
1 similar or slightly better avoidance oF Moire patterns. However, should operation of the random number generator assume a more regulàr pattern due to noise perturbations in the total machine system, it is conceivable that the dot patterns might possibly align themselves in some fashion, resulting in more pronounced Moire patterns.
The matrix shown in Figure 2A is repeated in Figure 6A. Employing the apparatus of Figures 4 and 5, random number generator 65 may cause initia-tion.of the pseudo halftone generator 40 such that any of the equivalent matrices of Figures 6B through 6D are produced for any given series of lines.
The matrix of Figure 2A is also repeated at Figure 7A. This illustrates that random number generator 65 and adders 77 through 79 may be recast to vary the row organization of the matrix and still produce a totally equivalent matrix.
By "equivalent" is meant that the matrices will give the same shape of dot pattern throughout an area of uniform gray scale level and will differ one from the other only at boundaries between different gray scale levels.
Figure 8 illustrates the weight distribution of the matrix of Figure 2A.
The gray scale matrix of Figure 2A is shown as the ~-matrix and the correspond-ing weight distribution is shown as the S-matrix. The S-matrix is a matrix of the sums of each corresponding T~matrix position with the adjacent right and lower ~-matrix positions. Thus, line 85 represents the sums of the gray scale level matrix count value positions 0, 4, 11 and 15 which comprise a total of 30.
Figure 9 illustrates the gray scale matrix of Figure 2A and the correspond-ing dot patterns for all gray scale levels greater than zero. As may be visualized with reference to Figure 9, the important properties of the (4x4) `i pseudo halftone matrix are: (a) the elements of the matrix in each row can be generated in a cyclic fashion; (b) when the output copy is rescanned by a (2x2) size aperture, the dot patterns remain invariant; and (c) the elements in the matrix, which represent different intensity levels, are uniformly dis-tributed. For printing with matrices other than (4x4), properties (a) and (c) can be nearly preserved and property (b) cannot be preserved.
The design of the cyclic code pseudo halftone matrix can be broken into -`- ` 1047935 1 two parts: (1) choosing the order by row of the print lines of the matrixand (2) assigning the inltial value of the cyclic sequence in each row to a suitable column in the matrix. Thus, given a (NxN) cyclic code matrix, there w;ll be N rows of cyclic elements. Let GN be an N vector whose elements con-tain the initial values of the cyclic sequence, i.e. GN(.) is taken from 0, 1, 2, ...(N-1). Then, let TG be the corresponding pseudo halftone printing matrix formed from expanding the initial values given in G into rows of cyclic sequences. Hence, for a given GN' there are many ways to form TG, because -there are many ways to choose the starting columns for the initial values of the cyclic sequences in all the rows. A way of avoiding Moire patterns is to attempt to equalize the sums of any two adjacent rows. Thus, for any given N, GN can be chosen as follows: -O
(N-l) , ~-1 '~'-'.. ' ",' -:
(N-3) .
GN ~ .
. .
i: `', N-(N-2) Since the average value from 0 to (N-1) ;S (N-1)~2, the choice of GN above .
ensures that the sum of any two rows of elements in TG will have approximately"
equal values. In other words, TG formed by expanding GN defined above ~.
;,.:- ... .. , . . ~ :
~1)9L793S
1 minimizes the maximum difference between the sums of two adjacent rows of elements in TG. Note that the sum of the elements in each row in TG is equal to N-l xoN ~ ~ K N
k=o where xO is the initial value of the cyclic sequence. Hence, for a given N, the sum of the elements in each row is governed by the ini$ial value xO. The arrangement of the elements in GN would then completely define the sum of the elements in each row of TG. It should be noted that, as far as minimizing the maximum difference between the sums of two rows of elements in TG is concerned, GN above is not unique and GN defined in the following would give the same result:
O ':
(N-21 ~' (N-4) 4 ~
: . :
GN ~
: '~' ' '" ,:
: (N-3 : ~ . 1 :,, ', ; ~ (N~
30 For a chosen GN, i:t is then necessary to find the column for starting the ~:
inltial or low value of the cyclic sequence of each row to form the resultant matrix TG. A natural suggestion is that the starting value should be separated '1`:
` ~4793S
1 approximately equidistant to each other in the printiny matrix by approximately the ~ distance apart, and i~ ~ is not an integer, using the next higher integer Yalue. However, as illustrated in Figure lo, a repeated shift of the same amount of the initial values of the cyclic sequence in the larger matrices could create a low frequency line pattern. The example illustrated in Figure 10 is a (9x9) matrix, wherein shifting the initial values of the cyclic elements in each row by 3 results in a very apparent low frequency line pat-tern and, possibly, in a visually recognizable Moire pattern.
Thus, in order to avoid the line pattern effect, the low values for the print lines from one row to the next are shifted by integer ( ~f~) and alter-nate print lines ~y the integer ( ~ *l) columns. Figure 11 illustrates exemplary cyclic code pseudo halftone printing matrices from a (5x5) matrix to a (12x12) matrix. Only the lowest or initial values of the cyclic sequences -;~
in the rows are shown. The other elements in each row can be obtained by add~
ing N to its preceding element in a cyclic fashion.
Figure 12 illustrates the parallel arrangement of pseudo halftone genera- ~;
tors for printing all of the print lines in a single matrix simultaneously.
Assuming that the supplied gray scale data is of a (4x4) matrix, the same gray scale input on line 11 is supplied to inputs 101 through 104 of pseudo half-tone generators 106 through 109. If the gray scale data were of a (2x2) matrix, then one set of gray scale data would be supplied on lines 110 to inputs 101 and 102 of pseudo halftone generators 106 and 107, while another set of gray scale data would be supplied on lines 111 to inputs 103 and 104 of pseudo half-tone generators 108 and 107.
Clock 52 is connected to each of the pseudo halftone generators by line 24, which is synchronized with the printing mechanism 46 as shown in Figure 4.
Line 51 from the line start indicator of the printer is connected to random -number generator 65 to cause the generator to supply its then randomly selected ~;
number on output lines 115. The lines 115 are connected to the initiated input of halftone generator 106 and to adders 117, 118 and 119. Pseudo halftone generators 106 through 109 are identical to pseudo halftone generator 40 in Figure 4 and as detailed in Figure 1. Adders 117, 118 and 119 are each ., ~. , . ., ............... :
1~4~935 identical to the corresponding adders 77, 78 and 79 in Figure 5. The pseudo halftone generators supply their individual serial outputs on lines 121 through 124 to the respective parallel print lines of printer 46.
In operation, a line start signal on line 51 operates the random number generator 65 to supply the selected number from those on matrix print line 30 in 2A onto lines 115. The number represented by the signals on lines 115 are supplied to initiate pseudo halftone generator 106 and are supplied to adders 117, 118 and 119. The adders each respectively add a predetermined value with-out carry to the selec$ed number as illustrated respectively at lines 31 through10 33 in Figure 2A and supply the total to initiate the pseudo halftone generators 107, 108, and 109. A clock signal may operate a buffer such as buffer 45 to supply gray scale data on lines 11 to inputs 101 through 104 of the pseudo halftone generators 106 through 109. The pseudo halftone generators then oper-ate in response to the gray scale data to supply or not supply print signals on the respective output lines 121 through 124. At the next clock signal on line 24 from clock 52, the pseudo halftone generators are each incremented without carry and compare the new count value to the applied gray scale level, once again supplying a print or not print signal on the respective outline lines 121 through 124. The incrementing and suitable application of gray 20 scale inputs continues until the end of the print line and start of the sub-sequent print line, at which time a line start signal is supplied on line 51. ~`
The repetition rate of ~he supply of gray scale data on line 11 as con-trolled by the clock signals on line 56 is related to the incrementing rate of the count values as controlled by the clock signals on line 56 by the number of print positions along the print line represented by a single gray scale level. Thus, where an entire (Nx~l) matrix area is represented by a single gray scale level, the ratio of the repetition rate to the incrementing rate is l/N.
Further, where only a (nxn) sub-matrix of the print matrix (NxN) is represented by a single gray scale level, the ratio of the repetition rate to ~he incre-30 menting rate is n/N.
While the invention has been particularly shown and described with refer-ence to preferred embodiments thereof, it will be understood by those skilled 7~35 1 in the art that the foregoing and other changes in form and details may be made therein without departing frnm the spirit and scope of the invention.
,:
, --- :
' ~ '' .
~ ' ~
" ~' ':
, "~
''`'~
~ ' :.' `:, `'. ': ....
"`' ~ ``' ':' "' .': ''' ' ' "
' ' '. .~ .
'~'.` ~ ' ~ ' '. ' ';' ' S~74-038. ~17~ ~
: , , `
trick the eye into believing that various shades of continuous tones are seen. Pseudo halftone images refers to the process of reproducing the con-tinuous tone images or gray scale with a printing device having a fixed ~ `
printed spot size and fixed spot-to-sPt spacing- The level of gray to be ~i reproduced thus is represented by a number of dots which are printed out of a specified line segment or matrix array of printing positions. If the individual spots are sufficiently small, they effect a merger insofar as the eye is concerned to ~orm a visual merger with the spaces between the dots to cause the eye to believe it is seeing various shades of continuous tones.
Electronics have been applied to image printing at various times with respect to the creation of half tone or pseudo half tone images. One example is Richard G. Sweet et al U.S. patent No. 3,373,437, issued May 14 1968 entitled "Fluid Droplet Recorder with a Plurality of Jets." This patent illustrates an ink jet printing mechanism wherein jets of ink are each caused to break into a uniform stream of droplets which are directed to , ~ , ~ 7935 1 impact the paper which ;s to be printed. Selected droplets wh;ch are not to be printed are charged as they break from the stream and are deflected by a constant electrostatic field to impact a gutter. A video intelligence sig-nal representing the relative brightness or inverse gray scale of the image to be reproduced ;s sampled and applied to the charging electrode into a R-C time constant circuit. The sampled pulse thus decays at a rate deter-mined by the R-C time constant. So long as the charging electrode stays above a predetermined threshold value, the droplets emanating fro~ the jet are charged and deflected to the gutter, leaving the paper unmarked. Upon the sig- -nal dropping below the threshold value, the rema;ning drops of the stream are uncharged and impact the paper. The ratio of drops impacting the paper with respect to the total number of drops produced prior to application of the next ~ -video signal comprises the gray scale level of the sampled video signal. In a printing line made up of a string of ink droplets or their unprinted posi- -tions, the droplets would thus be grouped periodically in accordance with the periods of the video signal, the droplets appearing towards the end of each period and the spaces at the beginning of each period.
Another example is shown in IBM Technical Disclosure Bulletin, "Pseudo Half Tone for Representing Continuous Tone Images in Black-While Facsimile ; -Systems," G~K. Machol, Vol. 9, No. 6, Nov. 1966, pages 636-637. This example illustrates the derivation of pseudo half tone matrices in accordance with a -~
predetermined logic. The analog Yideo si3nal is applied to a quantizer which samples the ~ideo signal and provides a binary coded output indicating the ~ ~ -gray scale level of the sample point. The logic circluit then transforms the binary coded values into a white zero or black one for each print position of `
the matrix. Also shown is the incorporation of an alternating bit in the logic to allow estimation of an eight-level quantization by a four bit matrix through the incorporation of an extra black dot in every other matrix. In U.S. Patent No. 3,604,846, issued Dec. 26, 1972, David Behane et al, entitled "Method and System for Reconstruction of Half-Tone Images," the matrix pat-terns are placed in a table in storage and are accessed in accordance with a quantized video density value.
. . . .~ , , . , ~
~4~47935 1 The difficulty with the dot bunching arrangement of Sweet et al is that it generates pronounced Moire patterns in any larger area of the same or similar gray scale levels. Judicious choice of the matrices of Machol or of Behane et al may reduce but do not eliminate ~he Moire patterns. They, how-ever, introduce the requirement for complex logic circuitry or for an exten-sive storage and accessing mechanism to transform the quantized gray level into the proper form.
5ummary of the Invention It is therefore an object of the present invention to employ a simpli-fied arrangement to generate pseudo halftone images which reduces Moire pat-tern effects.
Another object of the present ~nvention is to provide an even distribu-tion of spots for a glven gray scale leYel such that the resultant image may ~ -be scanned and reproduced without significant variation.
According to the present invention, a pseudo halftone generator is initiated at the beginning of each line of printing in accordance with a ran~
dom number generator. The pseudo halftone generator is arranged to count cyclicly from the initiated count, incrementing the count by a preselected number N for each print position and recycling at the square o~ that number.
For each prin~:position, the count is compared to the gray scale level input to generate a print signal upon the level being greater than the count. Fur-ther in accordance with the present invention, an adjacent group of print lines may be arranged in a preselected matrix format of dimension N controlled by the initiation so that the lowest or reference count value for each line in the matrix is different and is shifted from the lowest count value of the pre-ceding line in the matrix by the number of print positions equal to an inte-ger close to the square root of N.
A feature of the present invention is that it is operable with the presentation of gray scale levels at a different frequency than the recycling frequency of the pseudo halftone generator.
The foregoing and other objects, features and advantages of the inven-tion will be apparent from the following more particular description of the ~L04793~;
1 preferred embodlment of t'ne invention as illustra~ed ~n the accompanying draw-ings.
Brief Description of the Drawings FIGURE 1 is a block diagram of a pseudo halftone generator constructed in accordance with the invention.
FIGURE 2 is a diagrammatic represen~ation of gray scale matrices.
FIGURE 3 is a diagrammatic representation of the matrices of Figure with the appropriate dot patterns.
FIGURE 4 is a schematic block diagram of a scanning and printing system employing the pseudo halftone generator of Figure 1.
FIGURE 5 is a block diagram of alternative initiation circuitry for the system of Figure 4.
FIGURE 6 is a diagrammatic representation of a series of gray scale matrjces.
FIGURE 7 is a diagrammatic representation of another series of gray scale matrices. ~-FIGURE 8 is a diagrammatic representation of a gray scale matrix and its corresponding weight distribution matrix.
FIGURE 9 is a diagrammatic representation of a gray scale matrix and its corresponding dot pattern.
FIGURE 10 is a diagrammatic representation of an exemplary gray scale matrix showing the low count value print position in each line.
FIGURE 11 is a diagrammatic representation of a series of various size gray scale matrices and showing the low count value print position for each line in accordance with the invention.
FIGURE 12 is a schematic block diagram o~ a series of pseudo halftone generators arranged in parallel in accordance with ~he invention.
Detailed Description of the Invention Pseudo halftone as discussed herein refers to the process of reproducing continuous tone images or gray scales with a printing device having only two levels of gray, namely, black and white, and operating with a fixed printed spot size and fixed spot-to-spot spacing. An embodiment of the pseudo half-, ." -. .- - . . ,; . . ., , , - , - .
- 11)47935 1 tone generator of the present inventlon is illustrated in Figure 1. In the example shown, the quantized, b~nary coded gray scale level is supplied to digital comparator 10 on lines 11. A cyclicly generated binary coded count is supplied to the comparator 10 on lines 12 through 15. Such digital com-parators are well known in the art and may be arranged to provide an output on line 16 when the digital value on lines 11 exceeds that oF lines l2-15.
Thus, whenever the gray scale input level exceeds the count appearing on lines 12-15, comparator 10 supplies an output signal on line 16 to operate the printing mechanism to print a dot in the corresponding print position.
At the beginning of a print line, an initiation count is supplied on line 17.
The initiation count may be achieved by random number generation or in a pre-determined manner, both of which will be discussed hereinafter. The signals on line 17 are applied to flip-flops 18 and 19 and to input lines 14 and 15 of comparator 10. The flip-flops 18 and 19 are selectively set or reset in ac-cordance with the signal appearing at the inputs 20 and 21 thereof. The resul-tant output signals, if any, are supplied on lines 12 and 13 to comparator 10.
For example, a positive signal appearing on any one of the lines may repre-sent the presence of a binary bit, whereas a negative signal on any one of the lines may represent the absence of a binary b;t. For each print position in a printing line of the printing apparatus, a pulse is supplied on line 24.
The pulse causes flip-flop 19 to reverse its condition, for example, changing from the state of supplying a binary 1 on line 13 to the state of supplying a 0 on line 13. When switching from the 1 to the 0 state, flip-flop 19 fur-ther supplies a carry output pulse on llne 25 to flip-flop 18. The occurrence of a pulse on line 25 similarly causes flip-flop 18 to reverse state. Thus, the net effect of a pulse appearing on line 24 is to add the binary equivalent of 4 to the count appearing at the input lines 12-15 to comparator 10. Thus, N is equal to 4. In the example shown, the count appearing on lines 12-15 to comparator 10 may range from a low of 0 to a high value, represented by all ones, of 15. Inasmuch as no carry output is supplied from flip-flop 18 to a higher order input to comparator 10, the effect of a pulse on line 24 when both flip-flops 18 and 19 are in the 1 state is to reverse both flip-flops SA9-74-038 ~6-1 to the 0 state. This has the e-ffect of recycling the count value appearing at input 12-15 upon that value otherwise reaching a value of 16 or greater.
For example, if the coun~ represented by lines 12 ~hrough 15 were a binary 1 (1110), the appearance o~ a pulse on line 24 wlll operate both ~lip-flops 18 and 19 to the opposite state so that the resultant count appearing on lines 12-15 is the binary equiYalent of 2 (0010). The count supplied on lines 12-15 can thus be set to be incremented by a selected number of 4 and to recycle at the count of 16, which value is the square of the preselected incrementing number N.
The exemplary gray scale range represented b~ lines 11 sùpplied to com-parator 10 is from a level of zero (all white) to 16 (all black~. Five lines are therefore required to indicate the highest gray scale level of 16, which takes the form 10000. If 1t is desired to employ only 16 gray scale levels from 0 to 15, a logic cîrcuit may be added to automatically con~ert the 15 level (1111) to the 16 level (10000).
Referring to Figure 2, some print line matrices are illustrated. That of Figure 2A corresponds to a realized pattern produced by the circuitry of Figure 1 wherein each print line is one of the rows of the Figure. With respect to row 30, the circuitry would be initiated with a count of 0 and incremented by four for each subsequent print position, recycling to 0 rather than counting 16. Row 31 would be initiated at a count of 11 and similarly incremented by four for each print;position. Row 32 is seen initiated with a count of 1 and also incremented byv four for each print position. Lastly, row 33 is initiated with a count of 10 and similarly incremented. The number in each print position of the matrices thus indicates the gray scale level which must be exceeded in order for a dot to be printed at that print position.
The pattern illustrated in the matrices of Figure 2B is different and is simply sequential in nature and is not arranged to be produced by circuitry resembling that of Figure 1.
Figure 3 illustrates exemplary dot patterns resulting from the same gray level input as printed in accordance with the matrices shown ln Figure 2. In the dot pattern of Figure 3A, the gray level value of 4 supplied at input S~*74-038 -7-, . . .
,. , ~ -; ~. - , - :
- ~4793S
1 lines 11 to comparator 10 causes the comparator to provide an output at print position 0, but exceeds the count supplied on lines 12-15 in no other position on line 30. Similarly, the gray scale level of 4 exceeds only the count of 3 in line 31, exceeds only the supplied count of 1 in line 32 and exceeds only the count of 2 in line 33. The net effect of the apparatus of Figure 1 is to scatter the dots throughout the matrix and thus avoid bunch-ing and a tendancy to create Moire patterns.
The arrangement of print position gray scale levels shown in Figure 2B
is not in accordance with the present invention and results in dot patterns as shown in Figure 3B. This dot pattern in essence alters the gray level to a combination of a black line and white space. Other gray scale levels applied to the matrix of Figure 2B similarly result in dot patterns having bunching of the dots and in the creation of Moire patterns.
Referring to Figure 4, a system is illustrated which employs the pseudo halftone generator 40 shown in detail in Figure 1. The pseudo halftone gener-ator may be employed to convert gray scale levels for printing either from the output of a scanner 41 as quantized by quantizer 42 or from a computer system 43. A switch 44 may selectively supply the data from the scanner and quantizer or from the computer system to a buffer 45. The switch 44 may alternatively supply the quantized data from quantizer 42 to the computer system 43 for storage in the memory thereof. The computer system may then selectively transmit the quantized gray scale information from the memory to the buffer 45. Buffer 45 is required because the data transmission char-acteristics of the quantizer 42 or o~ the computer system 43 may not be identical to that of the pseudo halftone generator 40. The gray scale data is supplied by 6uffer 45 on lines 11 to the pseudo halftone generator 40. The output of the generator is supplied on line 16 to a printer 46. The printer 46 may comprise an ink jet printer, a cathode ray microfilm printer, a wire pr;nter or even an impact printer or typewriter printer fitted with special spot printing elements. All such printers having a mechanism such as line start indicator 47 to indicate the termination of printing of one line and the beginning of printing of another line. The line start indicator 47 is .
~047935 1 connected to buf~er 45 via line 48 and to a random number generator 50 via line 51. Random number generators normally comprise a recycling counting mechanism such as a binary counter which is stepped or sequenced by accumu-lated noise signals exceeding threshold levels and are unclocked. Thus, the number generated is random as compared to any of the machine cycles or clock-ing of the receiving mechanism. The signal appearing on line 51 from line start indicator operates the random number generator to supply its then ran-domly selected output number on lines 17 to the initiating input of pseudo halftone generator 40. Clock 52 supplies a pulse on line 24 for each print position of printer 46. Synchronism between clock 52 and the printer 46 is obtained by virtue of either feedback from the printer on line 53, or feed-forward ~rom the clock 52 to drive the operation of printer 46 via line 54.
The clock 52 is arranged to provide appropriate signals on line 56 to drive buffer 54 in accordance with the relative periodicity or frequency of gray scale inputs relative to the printing rate of printer 46. For example, the clock signals supplied on line 56 may be one-half or may be one-fourth the frequency of those supplied on line 24 to represent that the gray scale levels change once every two print positions of printer 46. Except for the pseudo halftone generator 40 described in Figure 1, all the elements of Figure 4 are individually well known in the art and need not be described in detail here.
In operation, scanner 41 and quantizer 42 or computer system 43 deliver gray scale data, via switch 44~ to buffer 45. Upon line start indicator 47 of printer 46 indicating the start of a line by means of a signal on line 483 buffer 45 delivers gray scale data on lines 11 under the control of clock signal appearing on line 56. The line start indicator signal appearing on line 51 also operates random number generator 50 to initiate the pseudo half-tone generator by signals on line 17. The pseudo halftone generator 40 responds by comparing the initial count to the gray scale level and supplying ~;
the pseudo halftone output on line 16 in serial form, advancing the count sup-plied to comparator 10 in accordance with the pulses supplied on line 24 from clock 52.
SA9-74-0~8 -9-` iL~4l7 9 3~j 1 The arrangement of Figure 5 comprises an alternative to the randomnumber generator 50 in Figure 4. With reference to Figure 2A, use of the random number generator 50 will result in random starting points for each of the lines 30 through 33 rather than the uniform matrix pattern shown in the figure. This means that each print line will begln with any randomly selected number from 0 to 15.
To implement a pre-formatted matrix, the apparatus of Figure 5 is required. There, the line start indicator signal on line 51 is supplied to the input of a ring or four phase clock circuit 59. Line 60 of ring 59 is connected to random number generator 65~ to gate circuit 66, and to the "load"
inputs of registers 67 through 69. Lines 61 through 63 are connected, res-pectively, to gate circuits 71 through 73. Random number generator 65 is the same as random number generator 50 in Figure 4 except that the lowest order stages of the counter are preset and the noise excitation is applied to the binary 4 (N3 stage. Thus, the random number generator 65 randomly selects from values 0, 4, 8 or 12. The generator 65 is connected via lines 75 to gate circuit 66 and to adders 77 through 79. Gate circuits 66 and 71-73 are connected to lines 17 of Figures ~ and 1.
In operation, assume that ring 59 is in the last state, number 3. A
subsequent line start indicator signal on line 51 operates the ring to the "zero" state, providing a similar pulse signal on line 60. This signal operates the random number generator 65 to provide a randomly selected num-ber on lines 75. As an example, the numbers which may be randomly selected may be 0, 4, 8 or 12 employed in line 30 of Figure 2A as discussed above.
The randomly selected number is supplied to gate circuit 66, and adders 77 through 79 Each adder adds a different predetermined value to the randomly selected number and supplies the resultant output, respectively, to registers 67 through 69. To accomplish the matrix shown in Figure 2A, adder 77 must add the value of 11 to the randomly selected number on line 75 without a carry. This means that any total of binary 16 or greater will appear as a value of 0 or greater. For example, should the selected number on line 75 be 8, the total from adder 77 w;th a carry would be a binary 19, but appears - , ~ ;~ . . -~C~47935 1 as an output of binary 3 without the carry. Similarly, adder 78 adds a Yalue of 1 to the same randomly selected number on line 75 and adder 79 adds a value of 10 to the same randomly selected number. The signal on line 60 operates gate 66 to transmit the randomly selected number to lines 17 to `thereby ;n;tiate the pseudo halftone generator 40. The signal on line 60 also operates the "load" inputs of registers 67 through 69 to load therein the outputs of the respective adders 77-79. At the next line start indicator signal on line 51, the ring advances to the "1" state, providing an output signal on line 61. This signal operates gate circuit 71 to transmit the total appearing in register 67 to lines 17 of the pseudo halftone generator 401 thereby initiating the generator for print line 31 in Figure 2A. The next line start indicator signal on line 51 operates the ring 59 to state "2" to provide an output signal on line 62. This signal operates gate 72 to transmit the total from register 68 to the initiate lines 17 of the half-tone generator 40. This causes the halftone generator 40 to produce the print line 32 of Figure 2A. Las~ly, the next line start indicator signal on ~` !
line 51 operates the ring to condition "3" to provide an output signal on line 63. This signal operates gate circuit 73 to provide the total from register 59 to the initiate input lines 17 of the halftone generator 40. The halftone generator then produces the print line 33 shown in Figure 2A. The next line start indicator signal on line 51 again operates the ring to state "O" to cause the generation of another randomly selected number. As an alter~
native, registers 67 through 69 may be eliminated and line 80 employed to connect line 51 directly to the random number generator 65. This causes selection of one of the random numbers in line 30 of Figure 2A for each line start indicator signal on line 51. Operation with the registers 67 through 69 and without line 80 causes the generation of a regular matrix pattern sëlected to avoid Moire ~atterns as much as possible as shown in Figure 2A.
Operation of the system with line 80 and without registers 67-69 allows con-siderable shifting of the print lines 31 through 33 individually with respect to print line 30. This may result in a more random relative location of the dot pattern such as shown in Figure 3A which, if truly random, results in a ` ,,.. . , - ,, .. ,, , . , ., . . , ~ .. . .
~47g3S
1 similar or slightly better avoidance oF Moire patterns. However, should operation of the random number generator assume a more regulàr pattern due to noise perturbations in the total machine system, it is conceivable that the dot patterns might possibly align themselves in some fashion, resulting in more pronounced Moire patterns.
The matrix shown in Figure 2A is repeated in Figure 6A. Employing the apparatus of Figures 4 and 5, random number generator 65 may cause initia-tion.of the pseudo halftone generator 40 such that any of the equivalent matrices of Figures 6B through 6D are produced for any given series of lines.
The matrix of Figure 2A is also repeated at Figure 7A. This illustrates that random number generator 65 and adders 77 through 79 may be recast to vary the row organization of the matrix and still produce a totally equivalent matrix.
By "equivalent" is meant that the matrices will give the same shape of dot pattern throughout an area of uniform gray scale level and will differ one from the other only at boundaries between different gray scale levels.
Figure 8 illustrates the weight distribution of the matrix of Figure 2A.
The gray scale matrix of Figure 2A is shown as the ~-matrix and the correspond-ing weight distribution is shown as the S-matrix. The S-matrix is a matrix of the sums of each corresponding T~matrix position with the adjacent right and lower ~-matrix positions. Thus, line 85 represents the sums of the gray scale level matrix count value positions 0, 4, 11 and 15 which comprise a total of 30.
Figure 9 illustrates the gray scale matrix of Figure 2A and the correspond-ing dot patterns for all gray scale levels greater than zero. As may be visualized with reference to Figure 9, the important properties of the (4x4) `i pseudo halftone matrix are: (a) the elements of the matrix in each row can be generated in a cyclic fashion; (b) when the output copy is rescanned by a (2x2) size aperture, the dot patterns remain invariant; and (c) the elements in the matrix, which represent different intensity levels, are uniformly dis-tributed. For printing with matrices other than (4x4), properties (a) and (c) can be nearly preserved and property (b) cannot be preserved.
The design of the cyclic code pseudo halftone matrix can be broken into -`- ` 1047935 1 two parts: (1) choosing the order by row of the print lines of the matrixand (2) assigning the inltial value of the cyclic sequence in each row to a suitable column in the matrix. Thus, given a (NxN) cyclic code matrix, there w;ll be N rows of cyclic elements. Let GN be an N vector whose elements con-tain the initial values of the cyclic sequence, i.e. GN(.) is taken from 0, 1, 2, ...(N-1). Then, let TG be the corresponding pseudo halftone printing matrix formed from expanding the initial values given in G into rows of cyclic sequences. Hence, for a given GN' there are many ways to form TG, because -there are many ways to choose the starting columns for the initial values of the cyclic sequences in all the rows. A way of avoiding Moire patterns is to attempt to equalize the sums of any two adjacent rows. Thus, for any given N, GN can be chosen as follows: -O
(N-l) , ~-1 '~'-'.. ' ",' -:
(N-3) .
GN ~ .
. .
i: `', N-(N-2) Since the average value from 0 to (N-1) ;S (N-1)~2, the choice of GN above .
ensures that the sum of any two rows of elements in TG will have approximately"
equal values. In other words, TG formed by expanding GN defined above ~.
;,.:- ... .. , . . ~ :
~1)9L793S
1 minimizes the maximum difference between the sums of two adjacent rows of elements in TG. Note that the sum of the elements in each row in TG is equal to N-l xoN ~ ~ K N
k=o where xO is the initial value of the cyclic sequence. Hence, for a given N, the sum of the elements in each row is governed by the ini$ial value xO. The arrangement of the elements in GN would then completely define the sum of the elements in each row of TG. It should be noted that, as far as minimizing the maximum difference between the sums of two rows of elements in TG is concerned, GN above is not unique and GN defined in the following would give the same result:
O ':
(N-21 ~' (N-4) 4 ~
: . :
GN ~
: '~' ' '" ,:
: (N-3 : ~ . 1 :,, ', ; ~ (N~
30 For a chosen GN, i:t is then necessary to find the column for starting the ~:
inltial or low value of the cyclic sequence of each row to form the resultant matrix TG. A natural suggestion is that the starting value should be separated '1`:
` ~4793S
1 approximately equidistant to each other in the printiny matrix by approximately the ~ distance apart, and i~ ~ is not an integer, using the next higher integer Yalue. However, as illustrated in Figure lo, a repeated shift of the same amount of the initial values of the cyclic sequence in the larger matrices could create a low frequency line pattern. The example illustrated in Figure 10 is a (9x9) matrix, wherein shifting the initial values of the cyclic elements in each row by 3 results in a very apparent low frequency line pat-tern and, possibly, in a visually recognizable Moire pattern.
Thus, in order to avoid the line pattern effect, the low values for the print lines from one row to the next are shifted by integer ( ~f~) and alter-nate print lines ~y the integer ( ~ *l) columns. Figure 11 illustrates exemplary cyclic code pseudo halftone printing matrices from a (5x5) matrix to a (12x12) matrix. Only the lowest or initial values of the cyclic sequences -;~
in the rows are shown. The other elements in each row can be obtained by add~
ing N to its preceding element in a cyclic fashion.
Figure 12 illustrates the parallel arrangement of pseudo halftone genera- ~;
tors for printing all of the print lines in a single matrix simultaneously.
Assuming that the supplied gray scale data is of a (4x4) matrix, the same gray scale input on line 11 is supplied to inputs 101 through 104 of pseudo half-tone generators 106 through 109. If the gray scale data were of a (2x2) matrix, then one set of gray scale data would be supplied on lines 110 to inputs 101 and 102 of pseudo halftone generators 106 and 107, while another set of gray scale data would be supplied on lines 111 to inputs 103 and 104 of pseudo half-tone generators 108 and 107.
Clock 52 is connected to each of the pseudo halftone generators by line 24, which is synchronized with the printing mechanism 46 as shown in Figure 4.
Line 51 from the line start indicator of the printer is connected to random -number generator 65 to cause the generator to supply its then randomly selected ~;
number on output lines 115. The lines 115 are connected to the initiated input of halftone generator 106 and to adders 117, 118 and 119. Pseudo halftone generators 106 through 109 are identical to pseudo halftone generator 40 in Figure 4 and as detailed in Figure 1. Adders 117, 118 and 119 are each ., ~. , . ., ............... :
1~4~935 identical to the corresponding adders 77, 78 and 79 in Figure 5. The pseudo halftone generators supply their individual serial outputs on lines 121 through 124 to the respective parallel print lines of printer 46.
In operation, a line start signal on line 51 operates the random number generator 65 to supply the selected number from those on matrix print line 30 in 2A onto lines 115. The number represented by the signals on lines 115 are supplied to initiate pseudo halftone generator 106 and are supplied to adders 117, 118 and 119. The adders each respectively add a predetermined value with-out carry to the selec$ed number as illustrated respectively at lines 31 through10 33 in Figure 2A and supply the total to initiate the pseudo halftone generators 107, 108, and 109. A clock signal may operate a buffer such as buffer 45 to supply gray scale data on lines 11 to inputs 101 through 104 of the pseudo halftone generators 106 through 109. The pseudo halftone generators then oper-ate in response to the gray scale data to supply or not supply print signals on the respective output lines 121 through 124. At the next clock signal on line 24 from clock 52, the pseudo halftone generators are each incremented without carry and compare the new count value to the applied gray scale level, once again supplying a print or not print signal on the respective outline lines 121 through 124. The incrementing and suitable application of gray 20 scale inputs continues until the end of the print line and start of the sub-sequent print line, at which time a line start signal is supplied on line 51. ~`
The repetition rate of ~he supply of gray scale data on line 11 as con-trolled by the clock signals on line 56 is related to the incrementing rate of the count values as controlled by the clock signals on line 56 by the number of print positions along the print line represented by a single gray scale level. Thus, where an entire (Nx~l) matrix area is represented by a single gray scale level, the ratio of the repetition rate to the incrementing rate is l/N.
Further, where only a (nxn) sub-matrix of the print matrix (NxN) is represented by a single gray scale level, the ratio of the repetition rate to ~he incre-30 menting rate is n/N.
While the invention has been particularly shown and described with refer-ence to preferred embodiments thereof, it will be understood by those skilled 7~35 1 in the art that the foregoing and other changes in form and details may be made therein without departing frnm the spirit and scope of the invention.
,:
, --- :
' ~ '' .
~ ' ~
" ~' ':
, "~
''`'~
~ ' :.' `:, `'. ': ....
"`' ~ ``' ':' "' .': ''' ' ' "
' ' '. .~ .
'~'.` ~ ' ~ ' '. ' ';' ' S~74-038. ~17~ ~
: , , `
Claims (26)
1. Apparatus for generation of pseudo halftone spot patterns in accordance with pre-established gray scale levels, comprising:
an input for supplying said pre-established gray scale levels;
an output;
counting means for incrementing a count by a preselected number N for each print position, recycling at the square of said number; and comparator means connected to said input and to said counting means for comparing said gray scale level to said count for each said print position and providing a print signal therefor at said output upon said level exceeding said count.
an input for supplying said pre-established gray scale levels;
an output;
counting means for incrementing a count by a preselected number N for each print position, recycling at the square of said number; and comparator means connected to said input and to said counting means for comparing said gray scale level to said count for each said print position and providing a print signal therefor at said output upon said level exceeding said count.
2. The apparatus of Claim 1 wherein said gray scale levels each represent an area greater than one said print position; and said input is arranged to supply said preestablished gray scale levels sequentially at a repetition rate lower than the incrementing rate of said counting means.
3. The apparatus of Claim 2 additionally including:
clocking means connected to said counting means for supplying clock signals at said incrementing rate to increment said counting means and connected to said input for supplying said gray scale levels at said repetition rate.
clocking means connected to said counting means for supplying clock signals at said incrementing rate to increment said counting means and connected to said input for supplying said gray scale levels at said repetition rate.
4. The apparatus of Claim 2 for generation of said spot patterns for series of print positions, each series forming a print line, and additionally includ-ing:
initiation means for initiating the count of said counting means at the beginning of each said print line.
initiation means for initiating the count of said counting means at the beginning of each said print line.
5. The apparatus of Claim 4:
wherein said output is connected to a printing means wherein said input includes a buffer;
wherein said initiation means is connected to said printing means to receive therefrom signals indicating the beginning of each said print line; and additionally including clock means connected to and synchronized with said printing means, connected to said counting means for supplying first clock sig-nals at said incrementing rate of increment said counting means, and connected to said buffer for supplying second clock signals to drive said buffer to sup-ply said gray scale levels at said repetition rate.
wherein said output is connected to a printing means wherein said input includes a buffer;
wherein said initiation means is connected to said printing means to receive therefrom signals indicating the beginning of each said print line; and additionally including clock means connected to and synchronized with said printing means, connected to said counting means for supplying first clock sig-nals at said incrementing rate of increment said counting means, and connected to said buffer for supplying second clock signals to drive said buffer to sup-ply said gray scale levels at said repetition rate.
6. The apparatus of Claim 4 wherein said initiation means additionally in-cludes a random number generator for randomly selecting said initiation count.
7. The apparatus of Claim 6 for generation of pseudo halftone patterns for adjacent groups of N said print lines to form (NxN) matrices of print position,wherein:
said initiation means is additionally arranged to randomly select said initiation count for one of said N print lines and to separately supply said randomly selected initiation count for each of the remainder of said N print lines, each differing from said randomly selected count by a different predeter-mined value.
said initiation means is additionally arranged to randomly select said initiation count for one of said N print lines and to separately supply said randomly selected initiation count for each of the remainder of said N print lines, each differing from said randomly selected count by a different predeter-mined value.
8. The apparatus of Claim 7:
wherein said output is connected to a printing means;
wherein said input includes a buffer;
wherein said initiation means is connected to said printing means to receive therefrom signals indicating the beginning of each said print line;
and additionally including clock means connected to and synchronized with said printing means, connected to said counting means for supplying first clock sig-nals at said incrementing rate to increment said counting means, and connected to said buffer for supplying second clock signals to drive said buffer to sup-ply said gray scale levels at said repetition rate.
wherein said output is connected to a printing means;
wherein said input includes a buffer;
wherein said initiation means is connected to said printing means to receive therefrom signals indicating the beginning of each said print line;
and additionally including clock means connected to and synchronized with said printing means, connected to said counting means for supplying first clock sig-nals at said incrementing rate to increment said counting means, and connected to said buffer for supplying second clock signals to drive said buffer to sup-ply said gray scale levels at said repetition rate.
9. The apparatus of Claim 7 including in parallel, a separate output, count-ing means, and comparator means for each of said N print lines.
10. The apparatus of Claim 9:
wherein said outputs are connected to a printing means;
wherein said input includes a buffer;
wherein said initiation means is connected to said printing means to receive therefrom signals indicating the beginning of said print lines; and additionally including clock means connected to and synchronized with said printing means, connected to all said counting means for supplying first clock signals at said incrementing rate to increment said counting means, and connected to said buffer for supplying second clock signals to drive said buf-fer to supply said gray scale levels at said repetition rate.
wherein said outputs are connected to a printing means;
wherein said input includes a buffer;
wherein said initiation means is connected to said printing means to receive therefrom signals indicating the beginning of said print lines; and additionally including clock means connected to and synchronized with said printing means, connected to all said counting means for supplying first clock signals at said incrementing rate to increment said counting means, and connected to said buffer for supplying second clock signals to drive said buf-fer to supply said gray scale levels at said repetition rate.
11. The apparatus of Claim 7 wherein said initiation means additionally in-cludes N-1 separate adder means for altering said randomly selected initiation count for each of said remaining print lines.
12. The apparatus of Claim 11 wherein said random number generator and said adder means are arranged to form said print lines in a predetermined matrix for-mat pattern wherein the lowest count value for each said print line is differ-ent, ranging in value from 0 to N-1 and arranged to minimize the difference between the sums of the count values for any two adjacent print lines, and is shifted along said print line from the lowest count value of the preceding print line in the matrix by the number of print positions equal to an integer close to the square root of N.
13. The apparatus of Claim 7 wherein:
said pre-established gray scale levels may range in value from 0 and N2 and said input supplies said pre-established gray scale levels in digital form.
said pre-established gray scale levels may range in value from 0 and N2 and said input supplies said pre-established gray scale levels in digital form.
14. The apparatus of Claim 13 wherein said predetermined gray scale levels are uniform over each said matrix.
15. The apparatus of Claim 10 wherein-said pre-established gray scale levels may range in value from 0 to N2 and each represent an area equal to one said matrix, said input supplies said pre-established gray scale levels in digital form; and said repetition rate of said second clock signals of said clock means is related to said rate of said first clock signals by the factor of 1/N.
16. The apparatus of Claim 10 wherein:
said pre-established gray scale levels may range in value from 0 to N2 and each represent an area equal to an (nxn) submatrix of said matrix, said input supplies gray scale levels for said submatrices in digital form; and said repetition rate of said second clock signals of said clock means is related to said rate of said first clock signals by the factor of n/N.
said pre-established gray scale levels may range in value from 0 to N2 and each represent an area equal to an (nxn) submatrix of said matrix, said input supplies gray scale levels for said submatrices in digital form; and said repetition rate of said second clock signals of said clock means is related to said rate of said first clock signals by the factor of n/N.
17. A method of generation of pseudo halftone spot patterns along a print line of print positions in accordance with pre-established gray scale levels, com-prising the steps of:
initiating a count for the print position at the beginning of said print line;
supplying said pre-established gray scale level for said print position comparing said gray scale level to said count for said print position;
signalling the printing of said print position upon said comparing step indicating said level is greater than said count, incrementing said count by a preselected number N, recycling at the square of said number, for the next sequential print position of said print line; and returning to said supplying step.
initiating a count for the print position at the beginning of said print line;
supplying said pre-established gray scale level for said print position comparing said gray scale level to said count for said print position;
signalling the printing of said print position upon said comparing step indicating said level is greater than said count, incrementing said count by a preselected number N, recycling at the square of said number, for the next sequential print position of said print line; and returning to said supplying step.
18. The method of Claim 17 wherein said pre-established gray scale levels may range in value from 0 to N2.
19. The method of Claim 18 wherein said pre-established gray scale levels each represent an area greater than one said position; and said supplying step comprises supplying said pre-established gray scale levels sequentially at a repetition rate lower than the incrementing rate of said incrementing step.
20. The method of Claim 19 wherein said initiation step additionally comprises randomly selecting said initiation count.
21. The method of Claim 18 for generation of pseudo halftone spot patterns for an adjacent group of N said print lines to form (NxN) matrices of print positions wherein:
said initiation step comprises randomly selecting said initiation count for one of said N print lines, and separately supplying initiation counts for each of the remainder of said N print lines, each differing from said randomly selected count by a different predetermined value.
said initiation step comprises randomly selecting said initiation count for one of said N print lines, and separately supplying initiation counts for each of the remainder of said N print lines, each differing from said randomly selected count by a different predetermined value.
22. The method of Claim 21 for providing print signals to an output means which signals the beginning of each said print line for generation of said pseudo halftone spot patterns wherein:
said preselected gray scale levels each represent an area greater than one said print position;
said initiation step is additionally responsive to each said beginning of print line signal, said supplying step additionally comprises supplying said pre-established gray scale levels sequentially at a repetition rate lower than the incrementing rate of said incrementing step; and said signalling step additionally comprises signalling said output means.
said preselected gray scale levels each represent an area greater than one said print position;
said initiation step is additionally responsive to each said beginning of print line signal, said supplying step additionally comprises supplying said pre-established gray scale levels sequentially at a repetition rate lower than the incrementing rate of said incrementing step; and said signalling step additionally comprises signalling said output means.
23. The method of Claim 21 for the parallel generation of said pseudo halftone spot patterns for said adjacent group of N print lines wherein:
said initiation step additionally comprises simultaneously and separately initiating each of said N print lines; and said supplying, comparing, signalling and incrementing steps are accomp-lished essentially in parallel for each of said N print lines.
said initiation step additionally comprises simultaneously and separately initiating each of said N print lines; and said supplying, comparing, signalling and incrementing steps are accomp-lished essentially in parallel for each of said N print lines.
24. The method of Claim 21 for additionally forming said N print lines in a pre-determined (NxN) matrix format wherein:
said initiation step additionally comprises randomly selecting said initiated count for said one print line from a set of N count values, all separated by a value of N, and separately supplying said remaining initiation counts in a pattern to form said matrix format, wherein the lowest count value for each said print line is different, ranging in value from 0 to N-1 and arranged to minimize the difference between the sums of the count values for any two adjacent print lines, and is shifted along said print line from the lowest count value of the preceding print line in the matrix by the number of print positions equal to an integer close to the square root of N.
said initiation step additionally comprises randomly selecting said initiated count for said one print line from a set of N count values, all separated by a value of N, and separately supplying said remaining initiation counts in a pattern to form said matrix format, wherein the lowest count value for each said print line is different, ranging in value from 0 to N-1 and arranged to minimize the difference between the sums of the count values for any two adjacent print lines, and is shifted along said print line from the lowest count value of the preceding print line in the matrix by the number of print positions equal to an integer close to the square root of N.
25. The method of Claim 21 wherein said preestablished gray scale levels each represent an area equal to one said matrix, and said supplying step comprises supplying said preestablished gray scale levels in digital form at a repetition rate synchronized with and related to the incrementing rate of said incrementing step by the factor of 1/N.
26. The method of Claim 21 wherein said pre-established gray scale levels each represent an area equal to an (nxn) submatrix of said matrix; and said supplying step comprises supplying said pre-established gray scale levels for said submatrices in digital form at a repetition rate synchronized with and related to the incrementing rate of said incrementing step by the fac-tor of n/N.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US05/542,530 US4032978A (en) | 1975-01-20 | 1975-01-20 | Pseudo halftone print generator and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA1047935A true CA1047935A (en) | 1979-02-06 |
Family
ID=24164225
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA241,587A Expired CA1047935A (en) | 1975-01-20 | 1975-12-09 | Pseudo halftone print generator and method |
Country Status (13)
| Country | Link |
|---|---|
| US (1) | US4032978A (en) |
| JP (1) | JPS5524634B2 (en) |
| BE (1) | BE836819A (en) |
| BR (1) | BR7600355A (en) |
| CA (1) | CA1047935A (en) |
| CH (1) | CH591335A5 (en) |
| DE (1) | DE2556565C3 (en) |
| ES (2) | ES444425A1 (en) |
| FR (1) | FR2298242A1 (en) |
| GB (1) | GB1528377A (en) |
| IT (1) | IT1058322B (en) |
| NL (1) | NL187143C (en) |
| SE (1) | SE412152B (en) |
Families Citing this family (72)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4084196A (en) * | 1977-01-31 | 1978-04-11 | Dacom, Inc. | Electronic half-tone generating means for facsimile reproduction system |
| US4122496A (en) * | 1977-02-23 | 1978-10-24 | International Business Machines Corporation | Apparatus and method for digital halftone reproduction |
| US4150400A (en) * | 1977-03-31 | 1979-04-17 | International Business Machines Corporation | Methods of a coarse-scan/fine-print character reproduction with compression |
| FR2386213A1 (en) * | 1977-03-31 | 1978-10-27 | Ibm | FINE RESOLUTION REPRODUCTION PROCESS FROM A COARSE RESOLUTION SCAN |
| JPS5821979B2 (en) * | 1977-05-02 | 1983-05-06 | 日本電信電話株式会社 | Image signal halftone processing method |
| US4149194A (en) * | 1977-07-07 | 1979-04-10 | Xerox Corporation | Variable angle electronic halftone screening |
| US4185304A (en) * | 1977-07-07 | 1980-01-22 | Xerox Corporation | Electronic halftone screening |
| US4279002A (en) * | 1978-04-24 | 1981-07-14 | Xerox Corporation | Adapter for raster output scanning printer |
| JPS5536896A (en) * | 1978-07-14 | 1980-03-14 | Information Int Inc | Preparing halfftone duplicate and system thereof |
| US4189754A (en) * | 1978-11-24 | 1980-02-19 | The Mead Corporation | Gray tone reproduction apparatus |
| US4468706A (en) * | 1979-02-13 | 1984-08-28 | Coulter Systems Corporation | Imaging by varying the placement of elements in the pixels |
| JPS55146582A (en) * | 1979-04-27 | 1980-11-14 | Matsushita Electric Ind Co Ltd | Image recording method |
| JPS55164868A (en) * | 1979-06-08 | 1980-12-22 | Mitsubishi Electric Corp | Image recorder into plastic cards |
| JPS5616284A (en) * | 1979-07-18 | 1981-02-17 | Sanyo Electric Co Ltd | Ink jet printer |
| JPS5619777A (en) * | 1979-07-25 | 1981-02-24 | Yokogawa Hokushin Electric Corp | Gradation recording device |
| DE2931098C2 (en) * | 1979-07-31 | 1983-01-05 | Dr.-Ing. Rudolf Hell Gmbh, 2300 Kiel | Process for the production of printing plates |
| JPS5576390A (en) * | 1979-09-03 | 1980-06-09 | Ricoh Kk | Dither image processing method |
| JPS5646364A (en) * | 1979-09-10 | 1981-04-27 | Xerox Corp | Digital convertion in white and black image |
| DE2943018C3 (en) * | 1979-10-24 | 1982-05-13 | Siemens AG, 1000 Berlin und 8000 München | Method and arrangement for displaying a halftone image |
| US4246614A (en) * | 1979-12-26 | 1981-01-20 | Xerox Corporation | Binary graphic printer system having an electronic screen with shift control suited for rescreening |
| DE3049369A1 (en) * | 1979-12-29 | 1981-09-17 | Canon K.K., Tokyo | IMAGE REPRODUCTION DEVICE |
| JPS56155986A (en) * | 1980-05-07 | 1981-12-02 | Kogyo Gijutsuin | Method of generating dot pattern |
| JPS5725771A (en) * | 1980-07-23 | 1982-02-10 | Canon Inc | Picture signal processor |
| US4328426A (en) * | 1980-08-04 | 1982-05-04 | Xerox Corporation | Filter for image pixels |
| US4328425A (en) * | 1980-08-04 | 1982-05-04 | Xerox Corporation | Filter for image pixels |
| US4415920A (en) * | 1980-09-19 | 1983-11-15 | Canon Kabushiki Kaisha | Image signal processing unit |
| JPS5762664A (en) * | 1980-10-03 | 1982-04-15 | Ricoh Co Ltd | Binary processing method for analog picture signal |
| DE3037774C2 (en) * | 1980-10-06 | 1982-06-16 | Siemens AG, 1000 Berlin und 8000 München | Method and arrangement for displaying multicolored halftone images |
| US4449150A (en) * | 1981-01-19 | 1984-05-15 | Ricoh Company, Ltd. | Method of processing medium tone picture |
| JPS57159172A (en) * | 1981-03-24 | 1982-10-01 | Konishiroku Photo Ind Co Ltd | Image processor |
| JPS57193173A (en) * | 1981-05-22 | 1982-11-27 | Fuji Photo Film Co Ltd | Picture scanning and recording method |
| GB2103449B (en) * | 1981-06-29 | 1985-05-30 | Nippon Telegraph & Telephone | Method and apparatus for gray level signal processing |
| US4706077A (en) * | 1981-09-08 | 1987-11-10 | Xerox Corporation | Halftoning implementation for interactive image editing |
| DE3171227D1 (en) * | 1981-09-12 | 1985-08-08 | Hell Rudolf Dr Ing Gmbh | Method for the production of printing plates by means of printing irregularly distributed points |
| JPS58150370A (en) * | 1982-03-02 | 1983-09-07 | Sony Corp | Producing system of gradation signal for printer |
| GB2120896B (en) * | 1982-04-06 | 1987-01-28 | Canon Kk | Half-tone facsimile system |
| US5666444A (en) * | 1982-04-06 | 1997-09-09 | Canon Kabushiki Kaisha | Image processing apparatus |
| JPS5921173A (en) * | 1982-07-27 | 1984-02-03 | Mitsubishi Electric Corp | Heat transfer for color recording system |
| JPS59101368A (en) * | 1982-12-01 | 1984-06-11 | Canon Inc | Calculator with printer |
| US4672432A (en) * | 1983-04-28 | 1987-06-09 | Canon Kabushiki Kaisha | Method for recording a color image using dots of colorants of different densities |
| KR870001840B1 (en) * | 1983-06-08 | 1987-10-15 | 미쓰비시덴기 가부시기가이샤 | Printer device of television receiver |
| DD218042A1 (en) * | 1983-10-20 | 1985-01-30 | Robotron Bueromasch | METHOD FOR PRODUCING HALFTONE PICTURES BY INK INJECTION |
| US4680596A (en) * | 1984-08-02 | 1987-07-14 | Metromedia Company | Method and apparatus for controlling ink-jet color printing heads |
| JP2539783B2 (en) * | 1984-08-02 | 1996-10-02 | メトロメディア・カンパニ− | Inkjet printing head controller |
| US4736254A (en) * | 1984-11-22 | 1988-04-05 | Matsushita Electric Industrial Co., Ltd. | Method and apparatus for generating pseudo-halftone dots by comparing gray scale values of an original with dither threshold values stored in cells of a matrix array divided into imaginary matrices of elemental areas each containing one cell |
| FR2587572B1 (en) * | 1985-03-20 | 1994-02-18 | Canon Kk | IMAGE PROCESSING APPARATUS AND METHOD |
| US4797687A (en) * | 1985-05-01 | 1989-01-10 | Burlington Industries, Inc. | Patterning effects with fluid jet applicator |
| US4849768A (en) * | 1985-05-01 | 1989-07-18 | Burlington Industries, Inc. | Printing random patterns with fluid jets |
| US5175807A (en) * | 1986-12-04 | 1992-12-29 | Quantel Limited | Video signal processing with added probabilistic dither |
| USH996H (en) | 1987-01-20 | 1991-11-05 | Recognition Equipment Incorporated | High resolution page image display system |
| US4823189A (en) * | 1987-05-08 | 1989-04-18 | Tektronix, Inc. | Method and apparatus for creating a dither pattern |
| US5107346A (en) * | 1988-10-14 | 1992-04-21 | Bowers Imaging Technologies, Inc. | Process for providing digital halftone images with random error diffusion |
| JPH02162871A (en) * | 1988-12-15 | 1990-06-22 | Dainippon Screen Mfg Co Ltd | Dot forming method |
| US5051841A (en) * | 1989-10-16 | 1991-09-24 | Bowers Imaging Technologies, Inc. | Process for providing digital halftone images with random error diffusion |
| JPH0371219U (en) * | 1989-11-02 | 1991-07-18 | ||
| JPH03104641U (en) * | 1990-01-30 | 1991-10-30 | ||
| US5257115A (en) * | 1990-10-10 | 1993-10-26 | Fuji Xerox Co., Ltd. | Image recording apparatus using pseudo tone generated images |
| 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 |
| US5384647A (en) * | 1991-06-10 | 1995-01-24 | Minolta Camera Kabushiki Kaisha | Image processing apparatus comprising means for judging image area of pseudo half-tone image |
| US5233366A (en) * | 1992-05-26 | 1993-08-03 | Xerox Corporation | Half-tone printing with thermal ink jet using random spot distribution in one dimension |
| 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 |
| US5969729A (en) * | 1994-05-27 | 1999-10-19 | Colorspan Corporation | Ink jet printer with artifact-reducing drive circuit |
| JPH0890822A (en) * | 1994-09-28 | 1996-04-09 | Sony Corp | Printing method and image printer |
| US6164847A (en) * | 1997-01-28 | 2000-12-26 | Agfa Corporation | Imaging parameter detection |
| US5912683A (en) * | 1997-08-25 | 1999-06-15 | Lexmark International, Inc. | Method of printing with an ink jet printer using an enhanced horizontal resolution |
| US5991513A (en) * | 1997-09-30 | 1999-11-23 | Levien; Raphael L | Method and apparatus for suppressing moire patterns |
| US6633412B1 (en) | 1999-03-26 | 2003-10-14 | Seiko Epson Corporation | Smoothness enhancement in laser printing through periodic modulation of halftone cell |
| DE19919624C2 (en) * | 1999-04-29 | 2001-03-01 | Oce Printing Systems Gmbh | Method, system and computer program for compressing and transmitting raster image data |
| JP2005524224A (en) * | 2002-04-26 | 2005-08-11 | コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ | Active matrix display device |
| EP1584182A2 (en) * | 2003-07-18 | 2005-10-12 | Photogram Technologies | Macro-pixel image rendering apparatus and associated methods |
| US7706635B2 (en) * | 2005-10-20 | 2010-04-27 | Marvell International Technology Ltd. | Methods and systems for image scaling |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3294896A (en) * | 1963-07-24 | 1966-12-27 | Bell Telephone Labor Inc | Digital encoder for facsimile transmission |
| DE2228414C2 (en) * | 1972-06-10 | 1982-05-13 | Philips Patentverwaltung Gmbh, 2000 Hamburg | Control of matrix printer for half tone reproductions - has signal generators and noise generator to control matrix printer in statistical mode |
| GB1448112A (en) * | 1973-02-13 | 1976-09-02 | Int Publishing Corp Ltd | Electronic screening |
-
1975
- 1975-01-20 US US05/542,530 patent/US4032978A/en not_active Expired - Lifetime
- 1975-10-15 GB GB42145/75A patent/GB1528377A/en not_active Expired
- 1975-12-09 CA CA241,587A patent/CA1047935A/en not_active Expired
- 1975-12-11 FR FR7538572A patent/FR2298242A1/en active Granted
- 1975-12-16 DE DE2556565A patent/DE2556565C3/en not_active Expired
- 1975-12-16 IT IT30347/75A patent/IT1058322B/en active
- 1975-12-18 BE BE162893A patent/BE836819A/en not_active IP Right Cessation
- 1975-12-19 JP JP15067975A patent/JPS5524634B2/ja not_active Expired
- 1975-12-29 CH CH1682675A patent/CH591335A5/xx not_active IP Right Cessation
-
1976
- 1976-01-19 SE SE7600470A patent/SE412152B/en not_active IP Right Cessation
- 1976-01-19 ES ES444425A patent/ES444425A1/en not_active Expired
- 1976-01-19 ES ES444424A patent/ES444424A1/en not_active Expired
- 1976-01-20 NL NLAANVRAGE7600518,A patent/NL187143C/en not_active IP Right Cessation
- 1976-01-21 BR BR7600355A patent/BR7600355A/en unknown
Also Published As
| Publication number | Publication date |
|---|---|
| ES444425A1 (en) | 1977-05-16 |
| CH591335A5 (en) | 1977-09-15 |
| BE836819A (en) | 1976-04-16 |
| DE2556565C3 (en) | 1980-10-23 |
| DE2556565B2 (en) | 1977-04-21 |
| FR2298242A1 (en) | 1976-08-13 |
| IT1058322B (en) | 1982-04-10 |
| DE2556565A1 (en) | 1976-07-22 |
| SE412152B (en) | 1980-02-18 |
| US4032978A (en) | 1977-06-28 |
| JPS5187921A (en) | 1976-07-31 |
| SE7600470L (en) | 1976-07-21 |
| NL7600518A (en) | 1976-07-22 |
| FR2298242B1 (en) | 1978-05-19 |
| NL187143C (en) | 1991-06-03 |
| ES444424A1 (en) | 1977-05-16 |
| GB1528377A (en) | 1978-10-11 |
| JPS5524634B2 (en) | 1980-06-30 |
| BR7600355A (en) | 1976-08-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA1047935A (en) | Pseudo halftone print generator and method | |
| US5150428A (en) | Method for generating halftone image data with randomly selected threshold array | |
| US4651287A (en) | Digital image processing algorithm for output devices with discrete halftone gray scale capability | |
| US4040094A (en) | Electronic screening | |
| GB1489433A (en) | Method for dot matrix recording | |
| ATE121582T1 (en) | ELECTRONIC DEVICE FOR HALFTONE GRAPHIC ARTWORK SUPPRESSING MOIRE PATTERNS USING PSEUDO-RANDOM CHARACTER SET SELECTION. | |
| US4701811A (en) | Tone reproduction method and apparatus | |
| EP0888683A1 (en) | Interpolation of greyscale levels | |
| US5140431A (en) | Digital electronic system for halftone printing | |
| GB2204976A (en) | Printer head driving | |
| JPH06122207A (en) | Halftone printing by thermal ink jetting using unidimensionally random dot distribution | |
| US4340912A (en) | Random screen generator apparatus for producing halftone images | |
| US20070058203A1 (en) | System and method for generating low visibility multi-bit halftone | |
| US5689344A (en) | Mitigation of tenting deletions in color xerographic printers | |
| WO1991010313A2 (en) | False density contour suppression using randomly modified input signals for comparison to threshold values | |
| JP3801364B2 (en) | Distributed PWM gradation processing apparatus and printer | |
| US5467120A (en) | Thermal transfer recording method and apparatus of both sublimation type and fusion type | |
| US6956676B2 (en) | Methods and systems for providing halftone screens | |
| JP3944711B2 (en) | Dither matrix generation method and printer | |
| US4945422A (en) | False density contour suppression using stored random probabilities to form print/no-print decisions | |
| JP2001111829A (en) | Dither generating device and its method and printer | |
| US6443549B1 (en) | Continuous tone reproduction using improved ink jet droplet dispersion techniques | |
| JPS572772A (en) | Thermal recording method | |
| JPS57135177A (en) | Dot matrix printer | |
| JPH0227151B2 (en) |