JP4168033B2 - Method for creating threshold matrix and threshold matrix - Google Patents

Description

  BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a threshold matrix creating method for creating a screen called an FM screen or a stochastic screen that converts a continuous tone image into a dot pattern that is a binary image, and the threshold matrix. Suitable for printing field devices such as plate setters, CTP (Computer to Plate) devices, CTC (Computer to Cylinder) devices, DDCP (Direct Digital Color Proof) systems, and other inkjet printers or electrophotographic printers The present invention relates to a method for creating a threshold matrix and the threshold matrix.

  Conventionally, in the printing field, an FM (frequency modulation) screen is employed in addition to a so-called AM (amplitude modulation) screen characterized by the number of lines, an angle, and a dot shape (see Patent Document 1).

  A technique for creating a threshold value matrix for the FM screen is proposed in Patent Document 1 described above.

  In the technique according to Patent Document 1, the threshold value matrix is created in ascending or descending order by determining the threshold position so that the distance between the already determined threshold position and the newly determined threshold position is the longest. The dot pattern of the binary image using the threshold value matrix created in this way has excellent characteristics that the dots are not shifted and the periodicity due to the repetition of the threshold value matrix does not appear.

  Regarding the creation of the threshold matrix, the following patent documents can be raised.

  The technology according to Patent Document 2 is a low-frequency component of a certain dot pattern, which includes a pixel having the weakest low-frequency component among white pixels (unblackened pixels) and a pixel having the strongest low-frequency component among blackened pixels. This is a technique for smoothing a dot pattern by extracting and replacing the extracted white pixels and blackened pixels (see Patent Document 2).

  The technique according to Patent Document 3 is a technique for determining a threshold position so that the next black pixel is arranged at the position where the low-frequency component of the threshold matrix is weakest in the threshold matrix (see Patent Document 3). ).

  In the technique according to Patent Document 4, when the threshold arrangement in the threshold matrix up to a certain gradation is determined, when determining the threshold position of the next gradation, the blackened pixels are arranged at positions where the low frequency component is not strengthened. In this technique, the threshold position is determined (see Patent Document 4).

  The technique according to Patent Document 5 is a technique for creating a threshold matrix by applying the techniques of Patent Documents 2 to 4 based on this dot pattern when an ideal dot pattern at a certain gradation is given. (See Patent Document 5).

JP-A-8-265 566 Japanese Patent No. 34003116 JP 2001-292317 A JP 2002-36895 A JP 2002-369005 A

  By the way, when the FM screen is used in offset printing, there is a problem that the roughness is conspicuous in the image quality. Also, the FM screen has problems such as increased dot gain and unstable image reproduction when printing, when outputting a film, which is an intermediate process of printing, or when outputting a printing plate using a CTP device. .

  In the FM screen according to the prior art, the dot size is such that the number of constituent pixels is one pixel, such as 1 (1 × 1) pixel FM, 4 (2 × 2) pixel FM, or four pixels. Is determined, the threshold value array constituting the threshold value matrix is determined by the algorithm for creating the FM screen to determine the output quality, and only the dot size is a parameter that determines the quality of the FM screen. For example, if the dot size is determined to be 3 × 3 pixels FM for an output system that cannot stably reproduce 2 × 2 pixel FM dots on the highlight side, halftone resolution (pattern frequency or pattern resolution) The image quality is reduced.

  FIG. 13 shows a dot pattern 1 on the highlight HL side with a dot percentage of 2 × 2 pixels FM with a dot percentage of 5% and a dot pattern 2 in a halftone with a dot percentage of 50%. In addition, the dot pattern 3 on the highlight HL side where the dot size is 3 × 3 pixels FM and the dot percentage is 5% and the dot pattern 4 in the halftone where the dot percentage is 50% are shown.

  14 shows a power diagram when FFT (Fast Fourier Transform) is applied to the dot pattern 2 of the 2 × 2 pixel FM of FIG. 13, and FIG. 15 shows the 3 × 3 pixel FM of FIG. The power figure at the time of applying FFT with respect to the dot pattern 4 is shown.

  In FIG. 13, the dot pattern 2 of the 2 × 2 pixel FM has less roughness than the dot pattern 4 of the 3 × 3 pixel FM at 50% of the halftone, but the reproducibility to halftone printing is poor. . On the other hand, in 50% of the halftone, the pattern frequency 6 of the dot pattern 4 of the 3 × 3 pixel FM is about 13 [c / mm], and is about the pattern frequency 5 of the dot pattern 2 of the 2 × 2 pixel FM. Lower than 20 [c / mm]. Here, the pattern frequencies 6 and 5 of the peak value are also called peak spatial frequencies fpeak, respectively.

  The output resolution of an output system such as an image setter and a CTP (Computer to Plate) apparatus (hereinafter, the resolution of the output system is referred to as output resolution R) is, for example, 2540 [pixel / inch] = 100 [pixel / mm] ( Or 2400 [pixel / inch] = 94.488 [pixel / mm]). In this case, the dot size of the 1 × 1 pixel FM is 10 [μm] × 10 [μm] (10. 6 [μm] × 10.6 [μm], and the dot size of the 2 × 2 pixel FM is 20 [μm] × 20 [μm] (21.2 [μm] × 21.2 [μm]). In this specification, the output resolution R is different from the pattern frequencies 5 and 6 of the dot patterns 2 and 4 shown in FIGS.

  The present invention has been made in consideration of the above-described problems of the FM screen according to the prior art, and is a threshold matrix that can reproduce a high-quality image that is optimal for an output system and excellent in printability. And a threshold matrix thereof.

The threshold matrix creation method according to the present invention is a threshold matrix creation method for converting a continuous tone image into a dot pattern, which is a binary image. In the threshold matrix creation method, the process of determining the size of the threshold matrix and the constituent pixels of the minimum size dot arrangement and determining the number of halftone, and determining the pattern frequency in a certain dot percentage, as in the dot percentage in front SL with the pattern frequency, the dot of the minimum size of each dot percentage When determining a candidate position to be generated, a white noise pattern having the same size as the threshold matrix is generated at a certain halftone percentage, the generated white noise pattern is converted into frequency domain data, and further filtered by a band filter of the pattern frequency Frequency domain data multiplied by the frequency domain data The data is converted into spatial region data, the value of each pixel of the spatial region data is compared with a pixel value corresponding to the certain halftone percentage by a comparator, a dot pattern of binary data is determined, and the dot of the binary data In the pattern, a process of determining a blackened area as a candidate position of the minimum size dot on the highlight side and a blank area as a candidate position of the minimum size dot on the shadow side ; A threshold matrix comprising a process of determining the number of dots of the minimum size to be newly set with the current half percent with respect to the half percent in which the dot pattern of the binary data is determined, and a dot with adjusted number of pixels In order to obtain a dot pattern corresponding to the size of the threshold, and sequentially determining the threshold position, and in the process of sequentially determining the threshold position After determining the threshold placement candidate position with reference to the candidate position for arranging the dot of the minimum size and the number of dots of the minimum size in each halftone percentage, the number of pixels constituting the dot is adjusted to match each halftone percentage. Then, the threshold arrangement position is determined (the invention according to claim 1).

  In this invention, the number of constituent pixels of the minimum size dot and the pattern frequency in the halftone of the dot pattern are set, and the dot pattern is created for each halftone based on the number of constituent pixels of the minimum size dot and the pattern frequency. In this case, the dot candidate positions are determined in the highlight area and the shadow area so as to be the pattern frequency, and the minimum size newly set with the current half percentage with respect to the half percentage where the dot pattern is determined. After determining the number of dots, the threshold is set so that the dot pattern can be generated by adjusting the number of dots to adjust to the percentage of each dot, and the output system is set in order. An optimal threshold value matrix can be created.

In this case, when the white noise pattern is converted to the frequency domain data, FFT is applied to the white noise pattern, and when the frequency domain data is converted to the spatial domain data, IFFT may be applied to the frequency domain data. (Invention of Claim 2).

  In addition, the halftone is set to a gradation of more than 10% and less than 90% in halftone (the invention according to claim 3).

  Further, in the process of determining the number of dots of the minimum size to be newly set by the current screen percentage, the ideal FM screen is set in the range from 0% to 50%, and from 0% to a predetermined screen percentage. The number of dots can be gradually decreased from the corresponding number of dots, and the number of increases can be determined to be a zero value in the range of the predetermined halftone to 50% (the invention according to claim 4).

  Further, in the process of determining the number of dots of the minimum size to be newly set by the current screen percentage, the ideal FM screen ranges from 0% to the first screen percentage in the range of 0% to 50%. In the range from the first halftone dot to the second halftone percent, the increase number is determined to be zero, and further from the second halftone percent to the range of 50%. Can be determined so as to increase gradually (the invention according to claim 5). By determining in this way, a predetermined pattern frequency can be obtained in halftone, and a dot pattern with a small dot gain can be created.

  In the invention according to claim 1 or 2, when adjusting the number of pixels, a low frequency component of a dot pattern formed by a threshold value matrix in which threshold values determined so far are arranged is extracted, and the number of pixels is determined. If the number of pixels is large, the threshold is set so that the strongest pixel of the low frequency component is removed from the existing dot. The arrangement position can be adjusted.

  Here, it is preferable to apply a visual characteristic filter when extracting the low frequency component. Furthermore, the extraction of low-frequency components is a density image (density dot pattern) in which a dot pattern reproduced on a recording medium such as a film, a printing plate, or paper is predicted by calculation based on the laser beam shape and the characteristics of the photosensitive material. It is also possible to convert this into a predicted density image (density dot pattern).

The threshold matrix of the present invention is an N × M (including N = M) matrix size threshold matrix that converts a continuous tone image into a dot pattern that is a binary image, and the output resolution of the output system is R [pixel / mm]. And the dot pattern pattern frequency generated by binarizing continuous tone image data in which each pixel value has a value corresponding to halftone percent 50 [%] is r [c / mm] A minimum of n (n is at least 1) constituent pixels until the number of dots corresponding to the pattern frequency r is in the vicinity of N × M / (R / r) ² [pieces] from halftone percent 0 [%]. A dot pattern is created so that dots of the size do not touch each other, and the net percentage after the area where the number of dots of the minimum size is in the vicinity of N × M / (R / r) ² in front Characterized in that it has a threshold array to create a dot pattern around a minimum size of a dot with a pixel without increasing the number of dots (the invention described in claim 6).

  According to the present invention, in the halftone, since the increase in the peripheral length is suppressed as compared with the conventional FM screen, the dot gain is reduced, which is optimal for the output system and excellent in printability. High quality images can be reproduced.

In this case, the output resolution of the output system is R [pixel / mm] in an N × M (including N = M) matrix size threshold matrix that converts a continuous tone image into a dot pattern that is a binary image. When the pattern frequency of a dot pattern generated by binarizing continuous tone image data in which each pixel value has a value corresponding to 50% of halftone is r [c / mm], halftone 0 [ %] Until the number of dots corresponding to the pattern frequency r is in the vicinity of N × M / (R / r) ² [pieces], the minimum size dot consisting of n (n is at least 1) constituent pixels Dot patterns attached so as not to touch each other are created, and in the net percentage after the area where the number of dots of the minimum size is in the vicinity of N × M / (R / r) ² , the existing minimum size Dot In addition to adjusting the area of the dots by attaching pixels around the area, if the existing dots touch, the new minimum size dots increase until the number of dots corresponding to the pattern frequency r is near. By having a threshold arrangement for creating such a dot pattern, the pattern frequency is substantially maintained in the halftone, and the dot pattern that can obtain a pattern frequency closer to the target pattern frequency in the halftone A threshold matrix that can be created is obtained.

  The threshold matrix creation method of the present invention sets a minimum size dot consisting of one or more predetermined constituent pixels, a pattern frequency in the halftone of the dot pattern, and a newly set number of minimum size dots in each halftone dot. Then, by sequentially determining the threshold arrangement positions under these restrictions, it is possible to create a threshold matrix that is optimal for the output system.

  By setting a minimum size dot consisting of a predetermined number of pixels, the image can be stably reproduced at the highlight part of the image, the output and printability can be improved, and a high-quality image can be reproduced with a percentage of the entire network. Can do. On the other hand, since the pattern frequency is not increased when a sufficiently fine pattern frequency is obtained, a threshold value matrix that generates a dot pattern with output suitability that does not increase the dot gain rapidly can be created. .

  More specifically, it is possible to create a threshold matrix that can generate an image with dots surely on the highlight side, with reduced roughness in halftones, and with a small dot gain.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 shows a basic configuration of a threshold matrix creation system 10 to which a threshold matrix creation method according to an embodiment of the present invention is applied.

  1 generates an arbitrary image data I including a test pattern of uniform density, and an image data generator for generating a two-dimensional address (x, y) of the image data I. 12 and a threshold value matrix storage unit 14 for storing a plurality of threshold value matrices TM for outputting a threshold value th read out at an address (x, y), and the threshold value th and the image data I are compared to output binary image data H. And a dot pattern generator 18 that generates dot pattern data Ha corresponding to the binary image data H output from the comparator 16, and the dot pattern represented by the dot pattern data Ha is a desired dot pattern. As described above, the threshold value matrix creating device 20 that determines the threshold value array (threshold position) of the threshold value matrix TM, and dot pattern data A dot pattern corresponding to Ha film, and an output system 22 for forming on a printing plate PP or printed material.

  Here, the threshold matrix storage unit 14 is a recording medium such as a hard disk, and the image data generator 12, the comparator 16, the dot pattern generator 18, and the threshold matrix generator 20 except for the output system 22 are personal computers ( Including an input device 20a such as a CPU, a memory, and a keyboard mouse, and an output device such as a display 20b and a printer 20c.) It can be constituted by function realizing means achieved by executing the program stored in the computer. . Moreover, the function realization means which comprises the threshold value matrix production apparatus 20 can also be comprised with a hardware. The configuration and operation of the function realizing means included in the threshold value matrix creating apparatus 20 will be described later.

  In this embodiment, the output system 22 is basically configured as a CTP apparatus having an exposure unit 26 and a drum 27 around which a printing plate material EM is wound, and is driven at high speed by a main scanning motor (not shown). Sub-scanning (not shown) of an exposure unit 26 that outputs a plurality of laser beams (recording beams) that are turned on and off for each pixel in accordance with the dot pattern data Ha with respect to the printing plate EM wound on the drum 27 rotated by the MS. By moving the drum 27 in the sub-scanning direction AS, which is the axial direction of the drum 27, a dot pattern which is a two-dimensional image as a latent image is formed on the printing plate material EM. Note that the number of laser beams may extend over several hundred channels.

  The printing plate material EM (usually four types of printing plate materials having different screen angles for CMYK) on which the latent image of the dot pattern is formed is developed by the automatic processor 28 and visualized. A printing plate PP in which is formed is prepared. The prepared printing plate PP is mounted on a printing machine (not shown), and ink is applied to the mounted printing plate PP.

  The photosensitive material included in the printing plate material EM includes an alkali-soluble binder and a substance that absorbs infrared light or near infrared light and generates heat as disclosed in Japanese Patent No. 3461377. The positive image recording material preferably contains a substance which is thermally decomposable and does not decompose so as to substantially reduce the solubility of the binder. The printing plate material for the printing plate PP is preferably an image recording material containing a photosensitive material in which a layer containing each of the above substances is provided on a support such as an aluminum plate or a polyester film.

  In this case, as the alkali-soluble binder, a material containing a phenol resin, an acrylic resin, or a polyurethane resin is used. In addition, as a substance that absorbs infrared light or near infrared light and generates heat, a dye, a pigment, or carbon black is used. Further, as the substance that is thermally decomposable and does not decompose, the substance containing onium salt, diazonium salt, or quinonediazide compound is used as the substance that substantially lowers the solubility of the binder.

  By transferring the ink applied to the printing plate PP onto a printing paper, which is a recording medium such as photographic paper, a desired printed matter having an image formed on the printing paper can be obtained.

  As will be described later, the output system is not limited to a scanning exposure apparatus using a so-called laser beam, but is an apparatus that draws a film, a printing plate, or a printed material by a surface exposure method or an inkjet method, and a CTC printer, etc. It can also be applied to.

  The threshold value array of the threshold value matrix TM stored in the threshold value matrix storage unit 14 is recorded on a portable recording medium such as a DVD, CDROM, CD-R, or semiconductor memory, and is portable. Is possible.

  Next, a threshold matrix creating method using the threshold matrix creating system 10 shown in FIG. 1 will be described with reference to the flowchart of FIG. Note that the main body that executes the program according to the flowchart of FIG.

  First, in step S1, three parameters are set. The first parameter is the size of the threshold matrix TM set in the threshold matrix storage unit 14, and the size N × N of the threshold matrix TM in which N × N thresholds corresponding to N × N pixels are stored. Set. In the threshold matrix TM, a threshold th that takes 0 to thmax is arranged at each position (element) determined by an address (x, y). The value of the maximum threshold thmax is set to “255” in a system having an 8-bit gradation, and “65535” in a system having a 16-bit gradation. Hereinafter, an example of the square threshold matrix size N × N will be described, but a rectangular threshold matrix size N × M can also be used. In practice, the threshold value matrix TM has a configuration in which a plurality of threshold value matrixes TM having a threshold value matrix size N × N having the same threshold value array are arranged in a tile shape corresponding to the size of an image (referred to as a super threshold value matrix STM). As used. The threshold th constituting the threshold matrix TM is determined in consideration of the threshold arrangement of the entire super threshold matrix STM.

  In this embodiment, the size of a pixel that can be output by the output system 22 is 10 [μm] × 10 [μm] = 1 × 1 pixel dot = 1 pixel. In this embodiment, 10 [μm] × 10 [μm] is a minimum unit that can be controlled by the exposure unit 26 when exposing and recording the printing plate material EM.

  The second parameter is the number of constituent pixels of the minimum size dot that can be stably output from the output system 22, in other words, can be stably formed on the printing plate PP output from the output system 22. In this case, the minimum size dot is 1 pixel dot (the number of constituent pixels of the minimum size dot is 1 pixel), 2 pixel dots, 3 pixel dots, 2 × 2 pixels (the number of constituent pixels of the minimum size dot is 4 pixels) ) Dots, 2 × 3 pixels (6 pixels) dots, 3 × 3 pixels (9 pixels) dots, and the like. In this embodiment, it is assumed that the minimum size dot that can be stably formed on the printing plate PP (in practice, a printed matter) is a 2 × 2 pixel dot composed of 4 pixels having a dot size of 2 × 2. .

  The third parameter is a pattern frequency at a predetermined halftone ratio, that is, a pattern frequency r of a halftone dot pattern in a halftone having a halftone (also referred to as density percentage) of 10% to 50%. The pattern frequency r of the halftone dot pattern designates the peak spatial frequency fpeak [c / mm] that the dot pattern has in halftone.

  This peak spatial frequency fpeak actually corresponds to reproduction of image details and affects image quality such as roughness. In this embodiment, the pattern frequency r is set to 20 [c / mm] fine enough visually, that is, 508 (20 × 25.4) [LPI: Line Per Inch] (fpeak = r = 20 [ c / mm]).

  Next, in step S2, the dot candidate position of the highlight HL and the dot candidate position of the shadow SD are determined so as to have the pattern frequency r in the halftone.

  In this case, first, as shown in FIG. 3A, a white noise pattern WH is created by the white noise generator 30 with a screen percentage of 50% of the same size N × N as the size N × N of the threshold matrix TM. The white noise pattern WH is an image in which 1 × 1 pixel dots are randomly arranged in a spatial area. The white noise pattern WH can be generated at an arbitrary value in the halftone of 10% to 90% halftone.

  Second, the white noise pattern WH is subjected to FFT (Fast Fourier Transform) by an FFT unit (Fast Fourier Transformer) 32, and further, a pattern frequency band filter (pattern frequency BPF) 34 is used to apply a pattern frequency r (± ΔΔ). ), A ring-shaped frequency domain data AFFT2 having a radius of the pattern frequency r is obtained as shown in FIG. 3B.

  Third, the frequency domain data AFFT is subjected to IFFT (fast inverse Fourier transform) by an IFFT unit (fast inverse Fourier transform) 36 to convert it into the spatial domain data A2 of the continuous tone image shown in FIG. 3C.

  Fourthly, the central gradation value (for example, 127 if the maximum gradation is 255) is compared by the comparator 38 against the value of each pixel in the spatial area data A2, and the binary data shown in FIG. A2_bin is created.

  In this binary data A2_bin, a blackened portion (region) is a dot candidate position in the highlight HL, and a white portion (region) is a dot candidate position in the shadow SD.

  The binary data A2_bin is a candidate position for arranging dots on the highlight HL side or the shadow SD side, and does not necessarily become a pattern of the binary data A2_bin when the dot percentage is 50%. This is because when the binary data A2_bin is not necessarily the optimum dot pattern of 50%, the degree of freedom is increased so that the optimum dot pattern is obtained.

  However, when it is desired to use a characteristic dot pattern at 50%, or when an optimum dot pattern of 50% can be obtained by correcting the dot pattern corresponding to the binary data A2_bin, the dot pattern of 50% Can be set.

  Next, in step S3, the minimum number of dots set newly with the current half percent with respect to the half percent where the dot pattern is determined (also referred to as the new minimum size dot number or the minimum size new dot number). Dn is determined. The minimum number of new dots Dn set in each halftone percentage P [%] is Dn (P) = Ds (P) −, where Ds (P) is the cumulative number of dots (cumulative value) in each halftone percentage P. Ds (P-1) [pieces].

  That is, in step S3, when the dot candidate positions are sequentially determined while increasing the halftone percentage, the current half percentage P is newly added to the previous half percentage P-1 in which the dot pattern has already been determined. The minimum number of dots Dn (P) to be set is determined.

  When the dot pattern has a halftone percentage P with respect to the threshold matrix size N × N, the total number of black pixels in the dot pattern corresponding to the size N × N of the threshold matrix TM is N × N × P / It is calculated as 100 [pieces]. If all the dots constituting the dot pattern are composed of only 2 × 2 (n = 4) pixel dots of the minimum size, the number of dots of the minimum size in each halftone dot P is Ds (P) = ( Since it is expressed by N × N × P / 100) / n, for example, N × N × P / 100 / n (n = 4) [pieces] as shown by the solid line na in FIG.

  At this time, the minimum number of dots Dn (P) newly set with the current halftone dot P is Dn (P) = Ds (P) −Ds (P−1) = (N × N / 100 / n). It becomes.

  Note that the vertical axis of FIG. 4 represents the cumulative value Ds in calculation of the newly set minimum number of dots (number of new dots) Dn. Actually, when the halftone percentage P is larger than 25 [%], adjacent dots of the minimum size come into contact with each other. Therefore, the number of dots actually formed in the dot pattern is the new dot shown in FIG. It becomes smaller than the cumulative value Ds of the number Dn.

  When the new dot number Dn is determined for each halftone of the new dot number accumulation characteristic na represented by a straight line, it becomes an FM screen according to the prior art, and at the time of printing or when outputting a film which is an intermediate process of printing Inconveniences such as increased dot gain and unstable image reproduction occur.

  Therefore, in one embodiment of the present invention, in consideration of the fact that the pattern frequency becomes smaller on the highlight HL side where the halftone percentage is less than 10%, all the dots are composed of the minimum size dots, and the halftone percentage is 10% to In the halftone area of 50%, the dots are thickened from the dots of the minimum size, and in this embodiment, the number of pixels of 5 (2 × 2 + 1) or more is used, and a new increase at each halftone percentage is performed. As shown in the new dot number cumulative characteristic nc represented by the dotted curve, for example, the number of dots Dn to be performed is such that the new dot number Dn of the minimum size to be set is about 10% to 25%. The number of new dots Dn is set to zero when the halftone percentage is from 25% to 50%. Alternatively, a new dot number accumulation characteristic nb represented by a one-dot chain line curve is set so that the new dot number Dn gradually increases again from 25% to 50%.

In this embodiment, the output resolution R of the output system 22 is 100 [pixel / mm] ← → 10 [μm / pixel], and the pattern frequency r of the halftone dot pattern is r = 20 [c / mm]. Therefore, if one side of N × N pixels is considered, the minimum blackened size dot consisting of 4 pixels per 100 pixels / mm (R pixels / mm) is 20 dots (1 dot is 2 × 2 pixels and r [C / mm]) [pieces] must be present. Accordingly, considering the size of N × N threshold matrixes TM, the cumulative value Ds of the number of new dots Dn up to halftone is (N / (R / r)) ² = N × N × (r / R) ² = N × N × (20/100) ² = N × N × 0.04 [pieces].

  With this setting, in halftones of 10% to 50% halftone, the total number of pixels constituting the dot pattern created by the threshold matrix TM in each halftone is the same as in the FM screen according to the prior art. That is, the halftone dot ratio is the same, but the number of dots is reduced. Therefore, the total perimeter of all dots constituting the dot pattern is reduced as compared with the FM screen according to the prior art. It will be.

  Here, as can be seen from the dot patterns 100 and 104 having the same area in FIGS. 5A and 5B shown as examples, the dot pattern 100 includes 16 1 × 1 pixel dots 102, and the perimeter is a dot pattern. 104 includes four 2 × 2 pixel dots 106, and the total area of the dots 102 constituting the dot pattern 100 and the total area of the dots 106 constituting the dot pattern 104 are the same. That is, although the dot percentages of the dot pattern 100 and the dot pattern 104 are the same, the dot pattern 100 is 16 [pieces] × 4 = the sum of the lengths of the black and white boundaries of the dot pattern per unit area, that is, the dot perimeter. It can be seen that the length is twice as long as the dot pattern 104 of 64, 4 [pieces] × 8 = 32.

  Accordingly, if the relationship of the cumulative value Ds of the dot number Dn to the halftone percentage is set as the new dot number cumulative characteristic nc, the increase in dot gain in the halftone is the cumulative value as the straight new dot number cumulative characteristic na. This can be suppressed compared to an FM screen in which Ds is set. Moreover, it has sufficient resolution in all halftone ranges and is equivalent to a conventional FM screen. However, if the new dot number Dn in the halftone is not increased as indicated by the new dot number accumulation characteristic nc (if it is too much reduced), each dot becomes larger and the roughness becomes easier to see and the image quality is improved. Since it falls, the pattern frequency of a dot pattern will become coarse.

  That is, even if the new dot number Dn is set for each halftone as in the new dot number accumulation characteristic nc, adjacent dots start to contact when the halftone percentage exceeds 25%, so the new dot number accumulation characteristic nc The cumulative value Ds of the set number of dots Dn is not reached.

  Therefore, in practice, as shown in the new dot number cumulative characteristic nb shown by the one-dot chain line curve in FIG. 4, the new dot number from when the halftone percentage exceeds 25% to 50% is set. A new dot number Dn of the minimum size is set so that Dn increases again by a substantially constant number. By setting the new dot number accumulation characteristic nb, it is possible to avoid the dots coming into contact with each other in the vicinity of 50% of the halftone, and to suppress the occurrence of tone jump.

  The cumulative value Ds of the number of new dots Dn exceeding the halftone percentage of 50% and up to 100% may be set to a curve symmetrical to the vertical line of the halftone percentage of 50%. It should be noted that the halftone percentage of 50% to 100% may be considered from the 100% side toward the 50% side, not the new number of dots Dn of blackened pixels, but new dots of white pixels (2 × 2 white pixels). Think in numbers.

  Next, the procedure for alternately determining the threshold value th in the ascending order and descending order from the highlight HL side and the shadow SD side in step S4 will be described with reference to the flowchart of FIG. In the following, in order to avoid complication, the procedure for sequentially determining the threshold th on the highlight HL side will be mainly described, but the same applies to the shadow SD side.

  Here, the threshold value th_hl {0 to (thmax-1) / 2} on the highlight side (0% to 50%) and the threshold value th_sd {thmax to (thmax-1) / 2 on the shadow side (100% to 50%) } Are determined as th_hl = 0 and th_sd = thmax in step S11, respectively.

  In the flowchart of FIG. 6, the arrangement order of the threshold th is determined in the order of threshold 0 → threshold thmax → threshold 1 → threshold thmax−1 →... → threshold (thmax−1) / 2. Arrangement positions (arrangements) of all the threshold values th are determined.

  When determining the arrangement (arrangement position) of the predetermined threshold th_hl on the highlight side, the dot center position is set in step S12. In this step S12, in the binary data A2_bin (FIG. 3D) determined in step S2, among the dot candidate positions in the highlight HL of the blackened portion (area), step S3 corresponding to halftone percentage. The dot center positions for the number of new dots Dn determined in (1) are set.

  As described in Japanese Patent Laid-Open No. 8-265666 (Patent Document 1), the center position of the dot is the threshold th_hl up to one gradation before the arrangement position of the threshold th in the threshold matrix TM has already been determined. For each dot that currently exists determined by −1, each dot that is set (attached) by the threshold th_hl for determining the arrangement position in the current threshold matrix TM is set to the most distant position. It is decided.

  For ease of understanding, a description will be given with reference to FIG. 7 schematically showing a super threshold value matrix STM in which nine threshold value matrixes TM1 to TM9 each having 25 threshold values are arranged. When determining threshold arrangement positions in ascending order from the highlight HL side of the threshold matrix TM or descending order from the shadow SD side, the central threshold matrix (5 × 5 threshold matrix in the example of FIG. 7) TM5 in FIG. Including the other threshold matrixes TM1 to TM4 and TM6 to TM9 having the same threshold value arrangement arranged around (near 9 in the example of FIG. 7) of the already determined threshold th (“1” in the example of FIG. 7). The center position of the newly arranged threshold th_hl is determined so that the distance between the arranged position and the arrangement position of the newly arranged threshold th_hl (“2” in the example of FIG. 7) is the longest. .

  In the example of FIG. 7, in the threshold value matrix TM5, the centrally surrounded threshold value “2” is concentrically enlarged at the same time around the four thickly surrounded threshold values “1”. Sometimes, the arrangement position including the point where the four circles contact or the position closest to the arrangement position and the blackened portion in the binary data A2_bin (FIG. 3D) is the arrangement position.

  More specifically, as shown in FIG. 8A, in the dot pattern 110 including the dots 108 with the threshold th determined to date, for example, the position 112 of the Δ mark is determined as the center position of the dot arrangement position. .

  Next, in step S13, a threshold arrangement position candidate (threshold candidate) th'_hl is set. In this case, 2 × 2 (n = 4) pixel dots, which are the dots of the minimum size determined in step S1 with the center position of the dot layout position determined in step S12 as the center, are set (placed), and a new threshold value is placed. Candidate, ie, threshold candidate th′_hl.

  Specifically, as shown in FIG. 8B, a dot pattern 114 is set in which a threshold candidate th′_hl of 2 × 2 pixel dots is set at the dot arrangement position 112 of Δ mark in FIG.

  Next, in step S14 to step S16, it is determined whether or not the total number of pixels of the dot pattern created by the threshold value matrix TM for which the arrangement of the threshold value th up to the present time corresponds corresponds to the current halftone percentage. To correct the total number of pixels. As the dot pattern, for example, continuous tone image data (image data I for generating a flat screen) corresponding to a halftone is generated from the image data generator 12 and stored in a threshold matrix. The comparator 16 compares the threshold value matrix TM in which the threshold values th to the threshold value th−1 that have been determined by the current time th stored in the unit 14 are arranged, and the resulting binary data H is obtained as a result. The dot pattern data Ha is obtained by being supplied to the dot pattern generator 18. A dot pattern based on the dot pattern data Ha is displayed on the display 20b or the like.

  Next, in step S14, the current pixel number th_hl_total obtained by adding the total number of pixels based on the thresholds 0 to th-1 whose arrangement positions have already been determined and the total number of pixels based on the newly set threshold candidates th′_hl is It is compared whether the number of necessary pixels required in the halftone is less than th_hl_num = N × N × th / thmax (th_hl_total <th_hl_num).

  If the number of pixels is small, it is necessary to increase the number of pixels (th_hl_num-th_hl_total) obtained by subtracting the current number of pixels th_hl_total from the required number of pixels th_hl_num. A new threshold candidate th ′ is set from a dot based on 0 to th−1 or a dot other than a new threshold candidate th′_hl whose arrangement position has not yet been determined.

  On the other hand, if there are many pixels, it is necessary to delete the required number of pixels th_hl_num−the number of current pixels th_hl_total, so in step S16, the dot from which to delete this pixel is selected from the dots based on the new threshold candidate th′_hl. And delete it.

  In step S16, among the dots constituting the dot pattern, there is a possibility that dots smaller than the minimum size dot are generated for several dots. In this embodiment, since the minimum size dot is a 2 × 2 pixel dot, the total number of pixels of the dot pattern created by the minimum size dot is a multiple of four. However, when adjusting the total number of dots to match halftone dots, 3 pixel dots, 2 pixel dots, or 1 pixel dot obtained by deleting 1 to 3 pixels from 2 × 2 pixel dots is required. There is.

  As proposed in Japanese Patent Laid-Open No. 2001-292317 (Patent Document 3), the processing in step S15 is performed by using dots 0 to th-1 whose arrangement positions have already been determined and new threshold candidates th ′. After the dot pattern (binary image data) on the spatial region composed of dots by _hl is subjected to FFT by the FFT unit 32 and converted to the frequency region, the high frequency is blocked by the LPF (low pass filter) 40, Further, the IFFT device 36 performs IFFT to return to the spatial region, and then performs processing for extracting low frequency components. Then, the position where the extracted low frequency component is the weakest may be set as the threshold candidate th ′ to which the pixel is to be added. However, if a 50% dot pattern is set in the process of step S2, the pixel is added to the blackened pixel of the 50% dot pattern and the low frequency component is the weakest. What is necessary is just to set to threshold value candidate th 'which should be.

  The extraction process of the weakest position of the low frequency component will be described in more detail. When the FFT is performed and converted into the frequency domain, the frequency component existing in the repetition frequency of the threshold matrix TM is a noise component (low frequency component). Therefore, in order to extract this low frequency component, the LPF 40 is applied.

  In this case, since the noise component is perceived by humans, when the high-frequency component is removed by the LPF 40, for example, at a spatial frequency of 0 [c / mm], a zero value and in the vicinity of a spatial frequency of 0.8 [c / mm]. A human visual characteristic filter 42 having a maximum sensitivity (assumed to be 1), a characteristic of about 0.4 at a spatial frequency of 2 [c / mm], and substantially zero at a spatial frequency of 6 to 8 [c / mm]. A low frequency component is extracted by applying the LPF as the LPF 40. As for the model of human visual frequency characteristics, the document "Design of minimum visual modulation halftone patterns" by the author J. Sullivan, L. Ray, and R. Miller, IEEE Trans. Syst. Man Cybern., Vol121, No.1. 33-38 (1991).

  Next, the IFFT unit 36 performs inverse Fourier transform on the low frequency component extracted by the LPF 40 to obtain a low frequency component on the spatial domain. The low frequency component has strength and weakness. The image composed of the low frequency component and the position of the threshold candidate th ′ in the threshold value matrix TM are compared on the spatial domain, and the low frequency component is the weakest (smallest value). ) The position is set as a threshold candidate th′_hl.

  In the case of the shadow SD side, the position where the low frequency component is the strongest (the largest value) may be set as the threshold candidate th′_sd.

  In step S16, the dot from which the pixel is deleted similarly extracts the low frequency component, and the pixel from the dot at the strongest (largest value) position of the low frequency component in the new threshold candidate th′_hl. Delete it. In the case of the shadow SD side, the pixel may be deleted from the dot by the new threshold th′_sd at the position where the low frequency component is the weakest (smallest value).

  FIG. 9A shows a dot pattern 120 according to this embodiment created in this way, with a minimum size of 2 × 2 pixel dots and a halftone dot percentage of 30%. FIG. 9C shows a dot pattern 122 of a one-pixel dot FM screen according to the prior art.

  The patterns obtained by applying the visual characteristic filter 42 to the dot patterns 120 and 122 as the LPF 40 and emphasizing the shading are shown as shading patterns 124 and 126 in FIGS. 9B and 9D. In addition, patterns in which the shading patterns 124 and 126 are represented by bird's-eye views are shown as bird's-eye patterns 128 and 130 in FIGS. 10A and 10B. In this bird's-eye view, the vertical axis represents 0 for white, 1.0 for black, and 0.3 for halftone dot 30. The horizontal axis is a pixel. As a result, it can be seen that the dot pattern 120 of FIG. 9A according to this embodiment is a pattern having a small amplitude with less variation in shading compared to the dot pattern 122 of FIG. 9C according to the prior art.

When setting the threshold candidate th′_hl in step S15 or step S16, as shown in Japanese Patent Laid-Open No. 2002-36895 (Patent Document 4 ), an IFFT is performed by the IFFT device 36 to obtain a spatial region. When the above low frequency component is obtained, the FFT unit 32 further performs the FFT, the specific frequency component extractor 44 extracts the specific frequency component in descending order of the intensity of the frequency component, and the extracted specific frequency component is the frequency. IFFT is applied in the order of increasing component intensity to form an image on the spatial region, and among these multiple images, the position with the weakest intensity component is selected as the threshold candidate th ′ or the threshold candidate th′_hl. Can also be set.

  Through the processes in steps S12 to S16 described above, a predetermined number of threshold values th can be set on the threshold value matrix TM corresponding to the positions where dots are newly added on the dot pattern.

  Next, in step S17, the dot pattern created by the determined threshold th is optimized. This dot pattern optimization process is an unnecessary process when a high-quality dot pattern is created by the processes up to step S16.

  This dot pattern optimization processing is performed by using either one or both of the technique disclosed in Japanese Patent No. 3400316 (Patent Document 2) and the technique disclosed in Japanese Patent Laid-Open No. 2002-369005 (Patent Document 5). Can be used.

  That is, according to the technique disclosed in Japanese Patent No. 340316, as described above, the low frequency component is extracted from the dot pattern created by the threshold th_hl, and the position having the strongest intensity among the extracted low frequency components. This is a process for reducing the intensity of the low-frequency component by replacing the pixels in the white area with white and replacing the pixels with the weakest intensity with black pixels. Here, the blackened pixel must be a pixel attached to the periphery of the dot (a pixel in contact with the periphery of the dot), and the threshold value th of the pixel is equal to the threshold value th of the dot.

  Further, according to the technique disclosed in Japanese Patent Application Laid-Open No. 2002-369005, the threshold value th is applied to the dot pattern created by the threshold value th as in the case of the aforementioned Japanese Patent Application Laid-Open No. 2002-36895 (Patent Document 2). After applying the FFT to the dot pattern created by the above, the visual characteristic filter 42 and the LPF 40 are applied, and further IFFT is applied to obtain a low frequency component on the spatial domain. Extract specific frequency components in descending order of component intensity, apply the IFFT to the extracted specific frequency components in descending order of frequency component intensity to create an image on the spatial domain, and do not enhance any of these multiple images By extracting and exchanging the pixel of the position where the intensity component is weakest and the pixel of the position where the intensity component is strongest among the positions, the intensity of the low frequency component is reduced. It is a process for. Also in this case, the extracted pixel must be a pixel attached to the periphery of the dot, and the threshold th of the pixel is equal to the threshold th of the dot.

In the low frequency component extraction process in steps S14 to S17, as disclosed in Japanese Patent Laid-Open No. 2002-369005 (Patent Document 5 ), a dot pattern output from an image output device corresponding to the dot pattern is disclosed. The density image corresponding to the above may be predicted and calculated (simulated) by the density image simulation unit (prediction unit) 46, and the low frequency component of the density image may be extracted. In this case, the density image simulation unit 46 actually outputs a test pattern from the output system 22 and measures how one dot of the original dot pattern is output on the grayscale image of the test pattern. Thus, the dot percentage in the density image close to the actual density image can be calculated from the dot pattern.

  For the density image, the exposure amount can be integrated and calculated from the beam shape of the laser light used in the output system 22, and the density image can be predicted from the gamma characteristics of the photosensitive material of the printing plate EM.

The calculation of the density image by calculation will be described in detail. First, a simulation for computer calculation of a laser beam for forming 1 × 1 pixel dots, 2 × 2 pixel dots,... On a recording medium such as a film F or the like. Determine the shape. The laser beam has a shape close to a Gaussian distribution, and can be substantially expressed by a beam diameter defined by the maximum value 1 / e ² of the amplitude value. An exposure amount for each dot is calculated from the laser beam and the dot pattern.

  Next, the calculated exposure amount for each dot, such as the calculated 1 × 1 pixel dot, 2 × 2 pixel dot,..., Is converted into the density of each dot by using the exposure characteristic of the photosensitive material such as a film, so-called gamma characteristic. To do. A density image (density simulation image) is obtained from the density of each dot thus obtained. A low frequency component can be extracted from the density image by the procedure using the above-described FFT. In practice, it is often possible to extract a low frequency component that is more effective for removing the noise component by extracting the low frequency component than the density image rather than extracting the low frequency component from the dot pattern.

  In this way, the position of the threshold th_hl in the threshold matrix is determined.

  Next, in step S18, the newly set threshold value th_hl is set to the threshold value th_hl + 1 of the next gradation (th_hl = th_hl + 1).

  Similarly, in step S22 to step S28, the threshold value th_sd on the shadow SD side is determined.

  In step S29, the threshold value th_hl determined from the highlight HL side is compared with the threshold value th_sd determined from the shadow SD side, and the threshold value th_hl and the threshold value are the same until the threshold value is equal to 50%. When th_sd is determined and becomes the same value, the creation of the threshold matrix ends.

  FIG. 11A to FIG. 11F show that the threshold value matrix TM thus created is finally generated by the dot pattern generator 18 by comparing with the grayscale continuous tone image data having the corresponding halftone by the comparator 16. The dot patterns 131 to 135 and 137 of a portion of the dot patterns of 10%, 20%, 30%, 40%, 50%, and 70% of the halftone dots are shown.

  The 70% dot pattern 137 may be a pattern obtained by reversing the black and white of the 30% dot pattern 133, or may be a pattern created independently.

  11A to 11F are created by selecting the new dot number cumulative characteristic nb represented by the dashed-dotted curve in the graph of FIG. 4, and up to the dot pattern 131 having a dot percentage of 10%. The dot pattern 131 is created with only 2 × 2 pixel dots of the minimum size dot. In the dot pattern 132 having a halftone dot percentage of 20%, the ratio of setting the 2 × 2 pixel dots of the minimum size dot is reduced, and pixels corresponding to the halftone dot are arranged around the existing dots (2 × 2 pixel dots). (4 to 12 pixel dots are present). With halftone percentages of 25% to 30%, new allocation of dots of the minimum size is not performed, and pixels are attached to existing dots to increase the blackening rate. Although the new allocation of dots of the minimum size is increased from about 35%, there is a function of forcibly connecting adjacent dots, so that the contact points between the dots can be dispersed. By setting in this way, a threshold matrix TM capable of generating a dot pattern with smooth gradation reproduction can be created.

  Thus, according to the above-described embodiment, the number of constituent pixels of the minimum size dot is determined, the pattern frequency r in the halftone of the dot pattern is determined (step S1), and further, based on the pattern frequency r, A dot candidate position is determined (step S2). Next, the minimum number of new dots Dc in the current half percent is set for the half percent where the dot pattern is determined (step S3). Then, by sequentially setting a threshold value th such that an optimum dot pattern is generated in each halftone percentage under the limitation of the number of new dots Dc of the minimum size dots and the halftone pattern frequency r (step S4), An optimum threshold value matrix TM for the output system 22 can be created. The threshold matrix TM optimum for the output system 22 means, for example, a threshold matrix TM which can generate an image with dots surely on the highlight side and with reduced roughness and a small dot gain on the halftone.

Further, according to the above-described embodiment, in the threshold matrix TM of N × M (including N = M) matrix size that converts a continuous tone image into a dot pattern that is a binary image, the threshold matrix TM is output. Dots generated by converting the continuous tone image data I having binary pixel data H with the output resolution of the system 22 being R [pixel / mm] and each pixel value having a value corresponding to 50 [%] of the halftone percentage P When the pattern frequency of the pattern data Ha is r [c / mm], the dot percentage corresponding to the pattern frequency r is N × M / (R / r) ² [pieces], since the halftone percentage P is P = 0 [%]. ] Until it becomes close, dot pattern data Ha is created so that dots of the minimum size consisting of n pixels (n is at least 1) do not touch each other, and the number of dots of the minimum size is N × / (R / r) in ² dot percentage after became neighboring region P, so as to have a threshold array to create a dot pattern data Ha to the ambient with the pixel does not increase the number of dots of the dot existing minimum size I have to.

In this case, in the threshold matrix TM, the dot area is obtained by adding pixels around the existing dots of the minimum size in the halftone percentage P after the area where the number of dots is in the vicinity of N × M / (R / r) ^2. Adjust.

  In the above description, one plate is described. However, when reproducing a color image, seven-color printing in which RGB colors are added to the separated CMYK colors, or CMYK colors + G colors + orange colors. Six-color printing is performed. In this case, m threshold matrixes having different threshold matrix sizes may be created for m (m> 4) colors. However, for complementary colors, the halftone percentage is hardly increased. Interference between complementary colors is small. Thus, the same threshold matrix can be used for complementary colors. For example, when CMYK color + RGB color inks are used, the same threshold matrix can be used for the M and G plates, the C and R plates, and the Y and B plates, respectively. Similarly, when the CMYK color + G color + orange color is used, the same threshold matrix can be used for the M and G plates, and the C and orange plates.

  The threshold value matrix created as described above is used, for example, as shown below.

  FIG. 12 shows a printing / plate making system 200 as an example in which the threshold matrix TM created by the threshold matrix creating apparatus 20 constituting the threshold matrix creating system 10 (see FIG. 1) is used.

  In the printing / plate making system 200, RGB image data captured by a digital camera 202 serving as an imaging device or RGB image data (or CMYK image data) captured by a plate making input device 204 serving as a scanner (image reading device) The RGB image data supplied to a RIP (raster image processor) 206 is temporarily converted into CMYK image data.

  In this case, the threshold matrix TM data (threshold matrix data) created by the threshold matrix creation device 20 is stored in the RIP 206 in advance on the hard disk of the RIP 206 through the optical disk 208 as a recording medium such as a CD-R or by communication. Has been.

  The RIP 206 compares the CMYK image data and the corresponding CMYK threshold matrix data, and converts them into CMYK dot pattern data (CMYK image data).

  The CMYK dot pattern data is sent to a so-called DDCP (also referred to as a real net proofer) 210 to create a print proof PRa on paper. This DDCP 210 confirms the presence or absence of noise components and the print quality before applying to the printing press 220. In this case, the printing paper itself may be used as the paper.

  Further, the CMYK dot pattern data is sent from the RIP 206 to the color ink jet printer 20c1 or the color electrophotographic printer 20c2, and the print proofs PRb and PRc can be easily created on the paper.

  Further, each CMYK dot pattern data is sent to an exposure unit 26 which is a film setter or plate setter constituting an output system 22 such as a CTC apparatus. When the exposure unit 26 is a film setter, a film F is created via an automatic processor 28, and this film F is overlaid with a printing plate material for a printing plate and exposed by a surface exposure device (not shown). A printing plate PP is created. When the exposure unit 26 is a plate setter as shown in FIG. 1, the printing plate PP is directly output through the automatic processor 28. A printing plate material EM or the like is supplied from a magazine 212 of photosensitive material (including printing plate material) to the exposure unit 26.

  Each CMYK plate PP is mounted on a plate cylinder (not shown) of the K plate printing unit 214K, the C plate printing unit 214C, the M plate printing unit 214M, and the Y plate printing unit 214Y constituting the printing machine 220. A color image is reproduced by overprinting the printing paper supplied from the printing paper supply unit 216 by the K plate printing unit 214K, the C plate printing unit 214C, the M plate printing unit 214M, and the Y plate printing unit 214Y. A printed matter PM is obtained. When the printing machine 220 has a CTC device configuration, CMYK dot pattern data is directly supplied from the RIP 206 by communication, and the printing plate material wound around the plate cylinder is exposed, recorded, developed, and directly printed. It is referred to as version PP.

1 is a basic configuration diagram of a threshold matrix creation system to which a threshold matrix creation method according to an embodiment of the present invention is applied. FIG. 2 is an overall flowchart for explaining a method of creating a threshold matrix by the system of FIG. 1 example; FIG. 3A is an explanatory diagram of a white noise pattern with halftone 50% created by 1 × 1 pixel dots. FIG. 3B is an explanatory diagram of FFT processing and band filter processing for a white noise pattern. FIG. 3C is an explanatory diagram in which the frequency domain image of FIG. 3B is converted into a spatial domain image obtained by IFFT processing. FIG. 3D is an explanatory diagram of a binarized image of the spatial region image of FIG. 3C. It is explanatory drawing of the number of dots with respect to halftone percent. FIG. 5A is an explanatory diagram for explaining the peripheral length of a small dot. FIG. 5B is an explanatory diagram for explaining the perimeter of large dots having the same halftone percentage as in FIG. 5A. FIG. 3 is a detailed flowchart for explaining the threshold arrangement position determination in step S <b> 4 in the entire flowchart shown in FIG. 2. It is explanatory drawing of the threshold value position determination process of the next gradation. FIG. 8A is an explanatory diagram of threshold candidate positions. FIG. 8B is an explanatory diagram in which dots of the minimum size are arranged at threshold candidate positions. FIG. 9A shows a dot pattern in which the minimum size dot is 2 × 2 pixel dots and the halftone dot percentage is 30%. FIG. 9B is a pattern diagram in which the light and shade of the dot pattern of FIG. FIG. 9C shows a dot pattern of a one-pixel dot FM screen according to the prior art. FIG. 9D is a pattern diagram that emphasizes the light and shade of the dot pattern of FIG. FIG. 10A is a bird's-eye view of the shading pattern of FIG. 9B. FIG. 10B is a bird's-eye view of the shading pattern of FIG. 9D. FIG. 11A is a diagram showing a dot pattern with a dot percentage of 10% generated by the threshold matrix according to this embodiment. FIG. 11B is a diagram showing a dot pattern with a halftone dot percentage of 20% generated by the threshold value matrix according to this embodiment. FIG. 11C is a diagram showing a dot pattern with a halftone dot percentage of 30% generated by the threshold value matrix according to this embodiment. FIG. 11D is a diagram showing a dot pattern with a halftone dot percentage of 40% generated by the threshold value matrix according to this embodiment. FIG. 11E is a diagram showing a dot pattern with a halftone dot percentage of 50% generated by the threshold value matrix according to this embodiment. FIG. 11F is a diagram showing a dot pattern with a halftone dot percentage of 70% generated by the threshold value matrix according to this embodiment. It is a block diagram which shows the printing and plate-making system as an example to which the threshold value matrix produced by the threshold value matrix production apparatus is applied. It is explanatory drawing which shows the dot pattern of 5% and 50% of the dot percentage of 2 * 2 pixel FM based on a prior art, and the dot pattern of 50% of the dot percentage of 3 * 3 pixel FM. FIG. 10 is a power diagram when FFT is applied to a dot pattern in which the dot percentage of 2 × 2 pixel FM is 50%. FIG. 10 is a power diagram when FFT is applied to a dot pattern in which the dot percentage of 3 × 3 pixel FM is 50%.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Threshold matrix preparation system 12 ... Image data generator 14 ... Threshold matrix storage part 16, 38 ... Comparator 18 ... Dot pattern generator 20 ... Threshold matrix preparation apparatus 22 ... Output system 26 ... Exposure unit 28 ... Automatic processor 30 ... White noise generator 32 ... FFT unit 34 ... Pattern frequency band filter 36 ... IFFT unit 40 ... LPF
42 ... Visual characteristic filter 200 ... Printing / plate making system

Claims (6)

In a method of creating a threshold matrix that converts a continuous tone image into a dot pattern that is a binary image,
Determining the size of the threshold matrix;
The process of determining the number of constituent pixels of the minimum size dot;
Of the halftones, the process of determining the pattern frequency with a certain net percentage ,
A white noise pattern having the same size as the threshold matrix size at the halftone percentage in determining the candidate position for arranging the minimum size dot at each halftone percentage so that the halftone percentage has the pattern frequency. The generated white noise pattern is converted into frequency domain data, and further, frequency domain data filtered by the band filter of the pattern frequency is obtained, the frequency domain data is converted into spatial domain data, and the spatial domain data is obtained. A pixel value corresponding to each halftone percentage is compared by a comparator to determine a binary data dot pattern, and a blackened region in the binary data dot pattern is defined as high. Area where white dots are used as candidate positions for the minimum size dot on the light side And determining a candidate location of the dots of the minimum size at the shadow side,
A step of determining the number of dots of the minimum size to be newly set with the current half percent with respect to the half percent where the dot pattern of the binary data is determined;
A step of sequentially determining the arrangement position of the threshold value so as to be a dot pattern corresponding to the size of the threshold value matrix composed of dots with adjusted number of pixels,
In the process of sequentially determining the arrangement position of the threshold,
After determining the threshold placement candidate position with reference to the candidate position for arranging the dot of the minimum size and the number of dots of the minimum size in each halftone percentage, the number of pixels constituting the dot is adjusted to match each halftone percentage. And determining a threshold arrangement position. The threshold value matrix creating method according to claim 1,
When converting the white noise pattern into the frequency domain data, the white noise pattern is subjected to FFT,
When converting the frequency domain data into the spatial domain data, the frequency domain data is subjected to IFFT.
A method of creating a threshold value matrix characterized by the above. The threshold value matrix creation method according to claim 1 or 2,
The halftone is a gradation of more than 10% and less than 90% in halftone percentage. In the creation method of the threshold value matrix of any one of Claims 1-3,
In the process of determining the number of dots of the minimum size to be newly set in the current screen percentage,
It is determined to gradually decrease from the number of dots corresponding to the ideal FM screen in the range from 0% to 50%, and in the range from 0% to the predetermined screen percentage, and in the range from the predetermined screen percentage to 50%, the increase number. The threshold value matrix is determined to be a zero value. In the creation method of the threshold value matrix of any one of Claims 1-3,
In the process of determining the number of dots of the minimum size to be newly set in the current screen percentage,
In the range of 0% to 50% of halftone dots, it is determined to gradually decrease from the number of dots corresponding to the ideal FM screen from the range of 0% to the first halftone percentage, and the first halftone to the second halftone. A method for creating a threshold matrix, characterized in that the number of increments is determined to be a zero value up to a range of percent, and is further determined to gradually increase from a second percent to a range of 50%. In a threshold matrix of N × M (including N = M) matrix size that converts a continuous tone image into a dot pattern that is a binary image,
The output resolution of the output system is R [pixel / mm],
When the pattern frequency of a dot pattern generated by binarizing continuous tone image data in which each pixel value has a value corresponding to halftone percent 50 [%] is r [c / mm]
A minimum of n (n is at least 1) constituent pixels until the number of dots corresponding to the pattern frequency r is in the vicinity of N × M / (R / r) ² [pieces] from halftone percent 0 [%]. A dot pattern is created with the size dots not touching each other,
Dots that do not increase the number of dots by adding pixels around the existing dots of the minimum size in the net percentage after the area where the number of dots of the minimum size is near N × M / (R / r) ² A threshold matrix characterized by having a threshold array for creating a pattern.

Images