Dither matrix generating method, dither matrix generating device, and program

Abstract

To generate a plurality of dither matrices that can effectively suppress image graininess without using complicated algorithms.SOLUTION: A generating method of a dither matrix includes: a step S2b of obtaining at least the same number of shift amounts as the number of dither matrices to be generated; a step S2c of selecting a set of the same number of shift amounts as the number of dither matrices to be generated from among the obtained shift amounts, performing shift processing based on each of the selected shift amounts, creating the same number of shifted dither matrices as the number of selected shift amounts, adding a reference dither matrix to the created dither matrix to form a dither matrix group, and obtaining a required number of dither matrix groups; and steps S2d and S2e of calculating an evaluation value indicating the degree of deterioration of graininess, selecting a dither matrix group with the smallest degree of deterioration, and generating dither matrices.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a dither matrix generating method, a dither matrix generating device, a program, and the like. [Background technology]

[0002] Methods for generating dither matrices are described in, for example, Patent Documents 1 and 2.

[0003] In

[0007] of Patent Document 1, it is stated that "since it is difficult to find an established method for generating a dither matrix, it basically relies on the experience and intuition of an expert."

[0004]

[0009] of Patent Document 1 states that "matrix elements are obtained using a matrix shift vector that indicates the shift direction and shift amount of the dither matrix."

[0005] In Patent Document 2, for example,

[0027] it is stated that "a temporary dither matrix is ​​obtained for each of the multiple colors of ink, and this temporary dither matrix is ​​obtained by shifting the reference dither matrix by a predetermined shift amount in the X direction and Y direction, and the shift amount for each ink is generated by a random number."

[0006] Furthermore, in

[0029] and

[0030] of Patent Document 2, it is stated that "using each provisional dither matrix, input image data is converted into provisional output image data, and, for example, 500 pieces of provisional output image data are obtained."

[0007] Furthermore, in

[0030] of Patent Document 2, it is stated that "an evaluation value E is calculated based on each provisional output image data. Furthermore, in

[0044] , with regard to the evaluation value, it is stated that "by introducing a new concept of evaluation value E (E=gs(n) / c(n)+gs(m) / c(m)), which is the sum of the value obtained by dividing the granularity gs(n) of a binary image when n types of ink are used by the incidence c(n) of ink dots when n types of ink are used, and the value obtained by dividing the granularity gs(m) of a binary image when m types of ink are used by the incidence c(m) of ink dots when m types of ink are used, it is possible to obtain an optimized shift amount, i.e., a dither matrix."

[0008] Furthermore, in

[0031] and

[0032] of Patent Document 2, it is stated that "the graininess gs(n) of a binarized image is calculated using the Shaw & Dooley algorithm, which predicts the graininess from the Wiener spectrum and the measured values ​​of the average density."

[0009] Furthermore, in

[0034] of Patent Document 2, it is stated that "the ink dot generation rate c(n) is expressed as c(n) = b(n) / a(n), where a(n) is the number of dots when n types of ink land without overlapping each other (i.e., the ideal number of dots), and b(n) is the number of dots when the landing positions of the n types of ink are shifted by a dither method using each temporary dither matrix (i.e., the number of dots that actually occur)." [Prior art documents] [Patent documents]

[0010] [Patent Document 1] Patent No. 4057515 [Patent Document 2] Japanese Patent Application Publication No. 2019-14167 Summary of the Invention [Problem to be solved by the invention]

[0011] Patent Document 1 only shows that the generation of a dither matrix basically depends largely on human experience, and that a matrix shift vector can be used to shift the dither matrix.

[0012] Patent Document 2 describes how "the shift amount is determined by a random number, a temporary dither matrix is ​​obtained by shifting a reference dither matrix, the image data is actually binarized to obtain temporary output image data, and an evaluation value is calculated using reinforcement learning, which is a form of machine learning, to determine the dither matrix."

[0013] However, it cannot be denied that, for example, the procedure becomes complicated because an image must actually be binarized, and that calculation of the evaluation value requires fairly complex processing.

[0014] Furthermore, as the number of ink colors used increases, the process of generating the dither matrix becomes more complex.

[0015] Furthermore, in binary images obtained using a dither matrix, for example, deterioration of graininess is easily noticeable in areas where the ink density value (ink value) is low, but Patent Document 2 does not mention a method for effectively reducing deterioration of graininess in this low density area.

[0016] An object of the present invention is to provide a dither matrix generating method that can generate a plurality of dither matrices that can effectively suppress the graininess of an image without using a complex algorithm or the like.

[0017] Other objects of the present invention will become apparent to those skilled in the art by referring to the following exemplary aspects and best modes, as well as the accompanying drawings. [Means for solving the problem]

[0018] In order to facilitate an understanding of the outline of the present invention, the following examples are given of embodiments according to the present invention.

[0019] In an aspect according to the present invention, a dither matrix generation method generates a plurality of dither matrices by shifting threshold values ​​of a reference dither matrix in at least one of a first direction and a second direction perpendicular to the first direction, the method comprising the steps of: acquiring a number of shift amounts at least equal to the number of the plurality of dither matrices to be generated; selecting, from the acquired shift amounts, sets of shift amounts equal to the number of the plurality of dither matrices to be generated, and performing shift processing based on each selected shift amount to create a number of shifted dither matrices equal to the number of selected shift amounts; adding the reference dither matrix to the created dither matrices to create a dither matrix group; thereafter updating the sets of shift amounts; and each time this is updated, updating the dither matrix group, thereby acquiring a required number of dither matrix groups; and calculating, for each of the acquired dither matrix groups, an evaluation value indicating the degree of deterioration in granularity when an output image is obtained by a dither method, selecting the dither matrix group with the smallest degree of deterioration in granularity, and generating the plurality of dither matrices using this method.

[0020] In this embodiment, the threshold conversion process sets the numerical values ​​of thresholds contained in the reference dither matrix that are equal to or greater than a predetermined value to 0, leaving only the small numerical thresholds related to the formation of low-density dots, and invalidating all other thresholds. Although it is normally impossible for a threshold to be "0," here, by forcibly replacing it with "0," it is possible to prevent it from affecting the calculation of the evaluation value.

[0021] For example, if the predetermined value is set to 4, the threshold values ​​included in the reference dither matrix will be only "0," "1," "2," and "3," which greatly simplifies the threshold values ​​and makes them easier to handle. Another advantage is that the evaluation value can be calculated by focusing on low-density dots where granularity is easily noticeable (in other words, where granularity is easily exacerbated).

[0022] When printing a light-colored image, many pixels are left unprinted, leaving the white of the media exposed. In this state, for example, if dots of different colors are closely spaced or overlapping, the graininess tends to worsen and image quality tends to deteriorate. Therefore, by focusing on the formation of low-density dots (corresponding to printing a light-colored image) and avoiding the close placement or overlapping of dots, graininess can be effectively improved. From this perspective, the above-mentioned method of calculating an evaluation value by focusing only on a small threshold value is effective.

[0023] Furthermore, if a dot is placed when the ink value of the image data is greater than a threshold value, then, for example, when the threshold value is "0," the minimum ink value for a dot to be formed is "1," and when the threshold value is "1," the minimum ink value for a dot to be formed is "2." In other words, the minimum ink value of a dot to be placed is uniquely determined in accordance with the threshold value, and taking this into consideration, it is possible to calculate an evaluation value by focusing on the threshold value itself.

[0024] Unlike the previously mentioned Cited Document 2, there is no need to take the trouble of using dither to obtain temporary output image data, so the effort involved in generating a dither matrix can be significantly reduced.

[0025] Next, for example, a plurality of shift amounts are generated, and a necessary number of shift amounts are selected from among them, and the reference dither matrix after threshold conversion is shifted according to each of the selected shift amounts.

[0026] For example, if three dither matrices (shift dither matrices) with different shift amounts are obtained, a total of four dither matrix groups are created, including the reference dither matrix. By updating the shift amount and performing the same process, the required number (multiple) of dither matrix groups can be created.

[0027] Next, for each dither matrix group, an evaluation value is calculated that indicates the degree of deterioration in granularity when an output image is obtained using the dither method.The dither matrix group that causes the least deterioration in granularity is then selected.The shifted dither matrix included in the selected dither matrix group is the desired dither matrix.

[0028] In this manner, in this embodiment, the evaluation value can be determined by repeating general calculations (addition, multiplication, etc.) by focusing only on the threshold values ​​of the reference dither matrix that have small numerical values, and no particularly complex algorithm is required.

[0029] For example, as in Patent Document 2 mentioned above, there is no need for complex data processing, such as calculating an evaluation value using reinforcement learning, a form of machine learning, based on the graininess of a binary image when multiple inks are used and the incidence rate of ink dots.

[0030] Therefore, according to this aspect, for example, it is possible to realize a dither matrix generation method that can generate a plurality of dither matrices that can effectively suppress the graininess of an image without using a complex algorithm or the like.

[0031] Furthermore, in this embodiment, a group of dither matrices capable of forming an image with less noticeable graininess can be selected through objective evaluation using evaluation values, which is also effective in automating dither matrix generation without relying on the experience of a skilled worker.

[0032] In an aspect according to the present invention, a dither matrix generating device generates a plurality of dither matrices by shifting threshold values ​​of a reference dither matrix in at least one direction of a first direction and a second direction orthogonal to the first direction, and includes a shift amount acquiring unit that determines shift amounts at least equal to the number of the plurality of dither matrices to be generated, and a shift processing unit that selects, from the acquired shift amounts, a set of shift amounts equal to the number of the plurality of dither matrices to be generated, and performs shift processing based on each selected shift amount, thereby generating a number of shifted dither matrices equal to the number of the selected shift amounts. and a dither matrix group creating unit that creates a required number of dither matrix groups by repeatedly performing a process of creating a dither matrix by adding the reference dither matrix to the created dither matrix and updating the set of shift amounts, and updating the dither matrix group each time the process is updated; and a dither matrix generating unit that calculates, for each of the created dither matrix groups, an evaluation value that indicates the degree of deterioration in granularity when an output image is obtained by a dither method, selects the dither matrix group that causes the minimum degree of deterioration in granularity, and generates the plurality of dither matrices using this evaluation value.

[0033] According to this aspect, it is possible to provide a dither matrix generating device that can generate a plurality of dither matrices that can effectively suppress the graininess of an image without using a complex algorithm or the like.

[0034] In an aspect according to the present invention, the program causes a computer to function as a shift amount acquisition unit, a dither matrix group creation unit, and a dither matrix generation unit.

[0035] According to this aspect, by realizing the functions of each part necessary for dither matrix generation using a software program (computer program), for example, the configuration of the dither matrix generation device is simplified and the implementation of the data processing unit is made easier.

[0036] Those skilled in the art will easily understand that the exemplified embodiments according to the present invention can be further modified without departing from the spirit of the present invention. [Brief explanation of the drawings]

[0037] [Figure 1] FIG. 1(A) is a diagram showing an example of the configuration of a dither matrix generating device (image processing device), FIG. 1(B) is a diagram showing an example of the configuration of a dither matrix group generating unit, and FIG. 1(C) is a diagram showing an example of the configuration of the main parts of an inkjet printer (printing device). [Figure 2] FIG. 2 is a flowchart showing an example of a procedure for generating a dither matrix according to a preferred embodiment of the present invention. [Figure 3] FIG. 3 is a flowchart showing an example of a specific processing procedure for the dither matrix group selection process (step S2) in FIG. [Figure 4] Figure 4(A) is a diagram showing an example of a reference dither matrix, Figure 4(B) is a diagram showing a specific example of threshold conversion processing, Figure 4(C) is a diagram for explaining threshold shifting and notation of shift amount, Figure 4(D) is a diagram showing the difference between a preferred shift amount (a shift amount that is selected) and an unpreferred shift amount (a shift amount that is not selected), and Figure 4(E) is a diagram showing the difference in graininess of the output image corresponding to each shift amount. [Figure 5] FIG. 5(A) is a diagram showing an example of a reference dither matrix, and FIGS. 5(B) to 5(D) are diagrams showing examples of dither matrices (shifted dither matrices) obtained by shifting the reference dither matrix. [Figure 6] FIG. 6 is a diagram showing more specifically the process of generating a dither matrix group using evaluation values. [Figure 7] FIG. 7 is a diagram for explaining statistical values ​​(statistical information) used to calculate the evaluation value. [Figure 8] FIG. 8 is a flowchart showing an example of a procedure for ranking dither matrices (an example of a procedure for determining the ranking number of high priority). [Figure 9]FIG. 9 is a diagram showing a specific example of a process for determining a combination of dither matrices with high priority. [Figure 10] FIG. 10 is a diagram specifically showing the procedure for the dither matrix ranking process. [Figure 11] FIG. 11 is a flowchart showing an example of a procedure for ranking dither matrices (a procedure following FIG. 8). [Figure 12] FIG. 12 is a diagram showing a specific example of the dither matrix ranking process shown in FIG. [Figure 13] FIG. 13 is a flowchart showing an example of the procedure for ink ranking processing and processing for associating ink with a dither matrix. [Figure 14] FIG. 14 is a diagram showing an example of an image output by the dither method. DETAILED DESCRIPTION OF THE INVENTION

[0038] The best mode described below is used to facilitate understanding of the present invention, and therefore, those skilled in the art should be aware that the present invention is not unduly limited by the embodiments described below.

[0039] Please refer to Fig. 1. Fig. 1(A) is a diagram showing an example of the configuration of a dither matrix generating device (image processing device), Fig. 1(B) is a diagram showing an example of the configuration of a dither matrix group generating unit, and Fig. 1(C) is a diagram showing an example of the configuration of the main parts of an inkjet printer (printing device).

[0040] As shown in FIG. 1A, a dither matrix generating device (image processing device) 100 includes a data processing unit 102, a control unit (processor) 110, and a storage unit (memory) 120.

[0041] The data processing unit 102 includes a reference dither matrix generating unit 10, a dither matrix generating unit 12, an ink-to-dither matrix association processing unit 30, and a storage unit (memory) 40.

[0042] The reference dither matrix generating unit 10 generates the reference dither matrix by, for example, the void and cluster method.

[0043] The dither matrix generating unit 12 also includes a dither matrix group generating unit 20 and a dither matrix ranking processing unit .

[0044] The generated data of the dither matrices associated with the inks (each of the inks of different colors) is stored in the storage unit 40. This data is loaded into the storage unit 214 of the input image processing device 210 in the inkjet printer (printing device) 200 shown in FIG. 1(C) and used.

[0045] 1(A) is loaded with a program PRG that causes the computer to function as the above-mentioned reference dither matrix generation unit 10, dither matrix generation unit 12 (including dither matrix group generation unit 20 and dither matrix ranking processing unit 28), and ink-to-dither matrix association processing unit 30. In other words, the program causes the computer to function as data processing unit 102.

[0046] By implementing the functions of each unit required for dither matrix generation using a software program (computer program), the configuration of dither matrix generating device 100 is simplified and the implementation of data processing unit 102 is facilitated.

[0047] 1(B), the dither matrix group generation unit 20 has a threshold conversion unit 21, a shift amount acquisition unit 22, a shift processing unit 23, a dither matrix group creation unit (dither matrix set creation unit) 24, a dither matrix group evaluation value calculation unit 25, a statistical value calculation unit 26, and a dither matrix group selection unit 27. The function of each unit will be described later.

[0048] 1(C) includes an input image processing device 210 and a print engine 220. The input image processing device 210 includes a screening processing unit 212 that uses a dither matrix, and a storage unit 214. The print engine 220 includes a head controller 222 and a print head 224.

[0049] Data for a plurality of dither matrices associated with inks (each of a plurality of inks of different colors) is loaded (stored) in advance in the storage unit 214 of the input image processing device 210.

[0050] The input image data is separated into ink colors, and the image data for each color is screened using the dither matrix corresponding to that color.

[0051] The output data obtained by the screening is transferred to a head controller 222 of a print engine 220, which drives a print head 224 based on the data to print an image on a medium (not shown).

[0052] Next, reference is made to Figure 2, which is a flowchart showing an example of a procedure for generating a dither matrix according to a preferred embodiment of the present invention.

[0053] In step S1, a reference dither matrix is ​​generated. In step S2, a process is performed to select the dither matrix with the smallest evaluation value from among the multiple dither matrices that have been generated.

[0054] This step S2 may include, for example, steps S2a to S2e. First, in step S2a, threshold conversion is performed. The threshold conversion process is a process of converting the numerical values ​​of some thresholds included in the reference dither matrix to other numerical values ​​(for example, "0"). Details will be described later.

[0055] In step S2b, a shift amount determination process is performed. Step S2b may include steps S2b1 and S2b2. In step S2b1, a shift amount is generated, for example, using a random number. In step S2b2, an appropriate shift amount is selected from the generated shift amounts. Since using shift amounts that are similar to each other can cause dots to overlap, shift amounts that are as dissimilar as possible are used. Details will be described later.

[0056] In step S2c, the reference dither matrix is ​​shifted and added to the shifted dither matrix to create a dither matrix group.

[0057] In step S2d, an evaluation value of the generated dither matrix group is calculated, and a statistical value is also calculated.

[0058] In step S2e, the smallest evaluation value is selected, and it is determined whether the evaluation value satisfies a predetermined condition based on statistical values. If it does, the evaluation value is adopted. If it does not, the evaluation value calculation process continues until the condition is satisfied. Then, a dither matrix group corresponding to the adopted evaluation value is selected. The content of the predetermined condition will be described later.

[0059] In step S3, a dither matrix ranking process is performed. Step S3 may include at least one of step S3a and step S3b.

[0060] In step S3a, evaluation values ​​are calculated for all combinations of dither matrices that are one number smaller than the predetermined number from among a predetermined number of dither matrices, the combination of dither matrices with the smallest evaluation value is identified, and the one dither matrix that is not included in the identified combination is assigned the lowest rank. Details will be described later.

[0061] In step S3b, a set of dither matrices is created by adding one dither matrix selected from a plurality of dither matrices whose priority is not specified to two or more dither matrices whose priority is specified, a set with the smallest evaluation value is specified, and a process is performed in which the added dither matrix included in the specified set of dither matrices is assigned the next priority level in the order of priority that has already been specified. Details will be described later.

[0062] In step S4, a process of associating the dither matrix with the ink is performed. Step S4 may include, for example, steps S4a and S4b.

[0063] In step S4a, a process for determining the order of inks is carried out. In step S4b, a process for associating each dither matrix with each ink is carried out. Details will be described later.

[0064] Next, reference will be made to Fig. 3. Fig. 3 is a flowchart showing an example of a specific processing procedure for the dither matrix group selection processing (step S2) in Fig. 2.

[0065] In step S2-1, threshold conversion processing of the reference dither matrix is ​​performed. The details of this processing will be described later. This processing is performed by the threshold conversion unit 21 in FIG. 1(B).

[0066] The value of the variable i is initially 1, and in this state, the process proceeds to step S2-2. In step S2-2, the shift amount is calculated randomly using, for example, random numbers.

[0067] In step S2-3, it is determined whether the shift amount is dissimilar to all of the shift amounts already adopted. The determination method will be described later. If the answer is N in step S2-3, the process returns to step S2-2, and if the answer is Y, the process proceeds to step S2-4. In step S2-4, the shift amount is adopted (decided) and the adopted (decided) shift amount is stored.

[0068] In step S2-5, it is determined whether the number of patterns of the adopted shift amount has reached c-1 (c: the number of shift dither matrices required). If the answer is N, the process returns to step S2-2, and if the answer is Y, the process proceeds to step S2-6. Steps S2-2 to S2-5 are performed by the shift amount acquisition unit 22 in FIG. 1(B).

[0069] In step S2-6, a dither matrix group (a set of dither matrices) Mi is created, which is composed of the reference dither matrix after threshold conversion processing and the dither matrix groups after shifts corresponding to the (C-1) shift amounts employed. This process is performed by the dither matrix group creation unit 24 in FIG. 1(B).

[0070] In step S2-7, an evaluation value Ei is calculated for the dither matrix group Mi created in step S2-6. The evaluation value can be considered an index for determining the similarity between the reference dither matrix and the shifted dither matrix. The method for generating the evaluation value will be described later. This process is performed by evaluation value calculation unit 25 in FIG. 1(B).

[0071] In step S2-8, statistical values ​​(for example, the mean μ and the standard deviation σ) are calculated based on the evaluation values ​​E1 to Ei. This process is performed by the statistical value calculation unit 26 in FIG.

[0072] In step S2-9, the dither matrix group D with the smallest evaluation value is identified from among the dither matrix groups M1 to Mi. However, since the determination that the evaluation value is the smallest is a relative determination made by comparing multiple evaluation values ​​and may be somewhat lacking in reliability, the dither matrix group is not selected (adopted) at this stage.

[0073] In step S2-10, it is determined whether the value of i exceeds a reference value (for example, a predetermined value), and it is determined whether the specified dither matrix group D satisfies a condition based on the statistical value acquired in step S2-8. Specific examples of the condition will be described later.

[0074] If the conditions based on the statistical values ​​are satisfied, it is guaranteed that the dither matrix group D identified in step S2-9 has sufficient reliability to be adopted. Therefore, if the answer is Y in step S2-10, the process proceeds to step S2-12, where the c dither matrix groups D that satisfy the predetermined conditions are selected (adopted) as the determined dither matrix group F. If the answer is N, the value of i is incremented by 1 in step S2-11, and the process returns to step S2-2. The processing of steps S2-9 to S2-12 is performed by dither matrix group selection unit 27 in FIG. 1(B).

[0075] Next, reference will be made to Fig. 4. Fig. 4(A) is a diagram showing an example of a reference dither matrix, Fig. 4(B) is a diagram showing a specific example of threshold conversion processing, Fig. 4(C) is a diagram for explaining threshold shifts and notation of shift amounts, Fig. 4(D) is a diagram showing the difference between preferred shift amounts (shift amounts that are selected) and unpreferred shift amounts (shift amounts that are not selected), and Fig. 4(E) is a diagram showing the difference in graininess of the output image corresponding to each shift amount.

[0076] As shown in Fig. 4(A), the reference dither matrix DM10 has a configuration in which m pixels (m is a natural number greater than or equal to 2) are arranged in the horizontal direction (first direction) and n pixels (n is a natural number greater than or equal to 2) are arranged in the vertical direction (second direction). In the example of Fig. 4(A), m = 4 and n = 3 are set, resulting in a dither matrix with 3 rows and 4 columns. A numerical value indicating a threshold value is written in each pixel.

[0077] In the following description, the "horizontal direction" may be referred to as the "first direction" or the "left-right direction." The "vertical direction" may be referred to as the "second direction" or the "up-down direction."

[0078] The dither matrix has a function of, for example, binarizing an image. Reference is now made to Fig. 14, which is a diagram showing an example of an image output by the dither method.

[0079] In the example of FIG. 14, the gradation value (density value) of the input image 500 is compared with the threshold value of the dither matrix DM, and if the "gradation value (ink value) ≧ threshold value," it is determined that a dot should be placed (black area), and if not, it is determined that a dot should not be placed (white area). However, the meaning of the threshold value and the determination formula can be set arbitrarily, and are not limited to this. In this embodiment, if the "gradation value (ink value) > threshold value," it is determined that a dot should be placed (black area), and if not, it is determined that a dot should not be placed (white area). In other words, the larger the threshold value, the higher the density, and the smaller the threshold value, the lower the density.

[0080] Returning to Figure 4, the explanation will continue. Threshold conversion processing is performed on the reference dither matrix DM10 in Figure 4(A). Figure 4(B) shows the reference dither matrix DM20 after threshold conversion. Here, threshold conversion processing is a process in which threshold values ​​contained in the reference dither matrix that are equal to or greater than a predetermined value are forcibly set to "0."

[0081] FIG. 4(B) is an example of conversion when the predetermined value is "4." By setting the threshold values ​​equal to or greater than the predetermined value (here, 4) to 0, it is possible to leave only the small threshold values ​​related to the formation of low-density dots, and invalidate the other threshold values. In other words, the threshold values ​​that indicate densities equal to or greater than the predetermined density are invalidated, and the evaluation value is calculated using only the threshold values ​​that indicate densities lower than the predetermined density. It is normally impossible for multiple pixels to have a threshold value of "0," but here, by forcibly replacing them with "0," it is possible to prevent this from affecting the calculation of the evaluation value. By calculating the evaluation value using the numerical value of the threshold value of the reference dither matrix that indicates a density lower than a predetermined density, it is possible to accommodate both cases where it is set such that "the smaller the threshold value, the lower the density" and cases where it is set such that "the larger the threshold value, the lower the density."

[0082] As can be seen from Figure 4(B), the threshold values ​​contained in the reference dither matrix DM20 after threshold conversion are only "0", "1", "2", and "3", which greatly simplifies the threshold values ​​and makes them easier to handle. Another advantage is that evaluation values ​​can be calculated focusing on low-density dots where granularity is easily noticeable (in other words, where granularity is easily exacerbated).

[0083] When printing a light-colored image, many pixels are left unprinted, leaving the white of the media exposed. In this state, if dots of different colors are closely spaced or overlapping, the graininess tends to worsen and image quality tends to deteriorate. Therefore, by focusing on the formation of low-density dots (corresponding to printing a light-colored image) and avoiding the close placement or overlapping of dots, graininess can be effectively improved. From this perspective, the above-mentioned method of calculating an evaluation value by focusing only on a small threshold value is effective.

[0084] Furthermore, if a dot is placed when the ink value of the image data is greater than a threshold value, then, for example, when the threshold value is "0," the minimum ink value for a dot to be formed is "1," and when the threshold value is "1," the minimum ink value for a dot to be formed is "2." In other words, the minimum ink value of a dot to be placed is uniquely determined in accordance with the threshold value, and taking this into consideration, it is possible to calculate an evaluation value by focusing on the threshold value itself.

[0085] Unlike the previously mentioned Cited Document 2, there is no need to take the trouble of using dither to obtain temporary output image data, so the effort involved in generating a dither matrix can be significantly reduced.

[0086] A shift process is performed to shift the thresholds of the reference dither matrix DM20 after threshold conversion in FIG. 4(B) in pixel units in at least one of the horizontal direction (first direction) and vertical direction (second direction, or up / down direction). The horizontal shift amount is a, and the vertical shift amount is b. Here, a is a natural number that satisfies 0≦a≦m, and b is a natural number that satisfies 0≦b≦n. The number of shift amounts obtained is at least the same as the number of dither matrices to be generated. In other words, the number of shift amounts obtained is the same as or greater than the number of dither matrices to be generated.

[0087] The shift amounts when performing the above shifts a and b are expressed as (a, b). For example, when a certain threshold is shifted horizontally (to the right), if that threshold is located at the right end of the matrix, the threshold is moved back to the left end of that row. In other words, a cyclic shift is performed in the horizontal direction. When a certain threshold is shifted vertically (downward), if that threshold is located at the bottom end of the matrix, the threshold is moved back to the top end of that column. In other words, a cyclic shift is also performed in the vertical direction.

[0088] a and b can be calculated (generated) randomly, for example, using a random number function. However, the calculated shift amount cannot be adopted unconditionally. If similar shift amounts are adopted, dots of different colors may be placed very close to each other or the dots may overlap. Therefore, shift amounts similar to shift amounts that have already been calculated are not adopted (not selected). In other words, a shift amount is selected.

[0089] A method for selecting the shift amount is shown in Fig. 4(D). When a shift of, for example, (2, 1) is applied to the reference dither matrix DM10, the shift process can be expressed as a process of moving the reference dither matrix DM10 two pixels to the right and one pixel downward, as shown in D-1 in Fig. 4(D). In other words, the shift process can be expressed as a process of moving the position of the reference dither matrix relative to the reference dither matrix itself.

[0090] In the example of D-1 in FIG. 4(D), a shift of (2, 1) results in a shifted dither matrix DM30 (hereinafter sometimes referred to as a shifted dither matrix). When point O and point O' are set at the upper left corners of the original reference dither matrix DM10 and the shifted dither matrix DM30, respectively, in the example of D-1, point O' is outside the circle of radius r centered on point O. In other words, the distance between points O and O' is greater than the reference value r. In this case, the shift amount (2, 1) can be determined to have low similarity and can therefore be selected (adopted). If the distance between points O and O' is the same as the reference value r, the similarity can be determined to be low and may be selected. In this way, the similarity of the shift amounts can be determined by comparing the distance between virtual points set at the same position in each dither matrix with a predetermined reference value.

[0091] On the other hand, in the example of D-2 in FIG. 4(D), the shift amount is (1, 0), and this shift results in a shifted dither matrix DM40. In this case, point O' is inside a circle of radius r centered on point O. In other words, the distance between points O and O' is smaller than the reference value r. Therefore, the shift amount (1, 0) is determined to have a high degree of similarity and is not selected (not adopted).

[0092] If we imagine an output image obtained by dithering using the reference dither matrix DM10 (labeled "reference DM" in the figure) shown in E-1 of Figure 4(E) and the rejected shift dither matrix DM40 shown in E-2, there is a high possibility that dots of different colors will be arranged closely together, as shown in E-4. If such close arrangement of dots occurs in a low-density image area, this can cause graininess to worsen.

[0093] On the other hand, if we assume that an output image is obtained using the dithering method using the reference dither matrix DM10 shown in E-1 and the adopted shift dither matrix DM30 shown in E-3, as shown in E-5, dots of different colors are positioned quite far apart from each other, and no particular problems arise.

[0094] 4(D) and (E) have been described, determining the similarity of the shift amount between the reference dither matrix DM10 and the shift dither matrix, but when generating multiple shift dither matrices, the similarity of the shift amount between the already adopted shift dither matrix and the newly calculated shift dither matrix is ​​also determined. Simply replace the reference dither matrix DM10 in Figures 4(D) and (E) with the already adopted shift dither matrix, and replace the shifted dither matrices DM30 and DM40 with the newly calculated shift dither matrices. The above determination makes it possible to adopt a set of a reference dither matrix and one or more shift dither matrices that have low similarity to each other.

[0095] Next, reference will be made to Fig. 5. Fig. 5(A) is a diagram showing an example of a reference dither matrix, and Figs. 5(B) to 5(D) are diagrams showing examples of dither matrices (shifted dither matrices) obtained by shifting the reference dither matrix.

[0096] By shifting the reference dither matrix DM20 after threshold conversion shown in Fig. 5(A) by (2, 1), (1, 1), and (1, 2), the shifted dither matrices DM50, DM60, and DM70 shown in Fig. 5(B) to Fig. 5(D) are obtained. By repeatedly shifting the reference dither matrix DM20 by different shift amounts, it is possible to create the required number of shifted dither matrices.

[0097] In the example of Fig. 5, three dither matrices (shift dither matrices) DM50 to DM70 with different shift amounts are created, and a reference dither matrix DM20 is added to these three to create a total of four dither matrix groups. This process is performed by the dither matrix group creation unit 24 in Fig. 1(B).

[0098] The shifted dither matrices DM50 to DM70 shown in FIG. 5 are dither matrices that would not normally be adopted (having a high degree of similarity in the shift amount) based on the method described in FIG. 4(D), but for the sake of convenience, hereinafter, the shifted dither matrices DM50 to DM70 are assumed to be dither matrices adopted in the method described in FIG. 4(D).

[0099] Next, reference will be made to Fig. 6. Fig. 6 is a diagram showing more specifically the process of generating a dither matrix group using evaluation values.

[0100] A-1 to A-7 in Fig. 6 show the process of calculating the evaluation value. The dither matrices DM20 to DM70 shown in A-1 to A-4 are the same as the dither matrices described above. The three shifted dither matrices DM50 to DM70 included in this dither matrix group have been shifted using the shift amounts selected by the method described in Fig. 4(D), and so have some effect in suppressing deterioration of granularity. However, this effect is not sufficient.

[0101] Therefore, in this embodiment, further evaluation is performed using evaluation values, and undesirable dither matrix groups are not adopted.

[0102] In other words, for each dither matrix group, an evaluation value is calculated that indicates the degree of deterioration in granularity when an output image is obtained using the dither method.The dither matrix group that causes the least degree of deterioration in granularity is then selected.The shifted dither matrix included in the selected dither matrix is ​​the desired dither matrix.

[0103] A more specific explanation follows. In A-5 of Fig. 6, the threshold values ​​of pixels at the same position in each of the dither matrices DM20 to DM70 shown in A-1 to A-4 are summed (added). For example, if we focus on the pixel in the upper left corner of each dither matrix, we get "0 + 3 + 0 + 0 = 3." This total value is the numerical value that forms the basis of the evaluation value.

[0104] For the sake of convenience, let us assume two dither matrices, and say that the threshold value of a certain pixel in one dither matrix is ​​"1." The threshold value of a pixel in the other dither matrix that is in the same position as the pixel mentioned above (referred to as the "same pixel") will be either "0," "1," "2," or "3" because threshold conversion has been performed.

[0105] The combination of threshold values ​​for the same pixel in two dither matrices is either "1 and 0," "1 and 1," "1 and 2," or "1 and 3." We will examine what difference occurs in the degree to which the graininess of the output image using the dither method deteriorates in this case.

[0106] A threshold value of "0" means that the pixel has been determined not to be involved in the formation of low-density dots in the threshold conversion, except when the threshold value was "0" before the threshold conversion. Therefore, it can be determined that a "0" in a combination of "1 and 0" has little effect on the formation of low-density dots due to the threshold value of "1" in the other dither matrix.

[0107] The combination of "1 and 1" is determined to affect each other because both thresholds are related to the formation of low-density dots. The combination of "1 and 2" is similar, but as the threshold increases, the density of the dots that are printed increases, making it easier for graininess to worsen, so the degree of influence is determined to be greater. The combination of "1 and 3" is determined to have an even greater degree of influence.

[0108] From this perspective, when combining threshold values ​​for the same pixel, "0" has little effect on the deterioration of graininess due to low-density dots, and the degree of effect increases as the value increases, such as "1," "2," and "3." Therefore, adding up the threshold values ​​for the same pixel (pixel in the same position) in each dither matrix that makes up the dither matrix group provides an index value that indicates the degree of deterioration of graininess.

[0109] Basically, the index values ​​obtained in this way for each pixel are added up for all pixels to obtain an evaluation value for the dither matrix group (in other words, an index value indicating the degree to which granularity deteriorates).

[0110] However, if the index values ​​for each pixel calculated using the threshold values ​​as they are are added up for all pixels, the same evaluation value will be obtained for all dither matrix groups. In other words, the characteristics of the index value array are not reflected in the evaluation value. Also, because the numerical values ​​indicating the dither matrix thresholds are limited to small values, the range of the evaluation value is limited, and even when comparing evaluation values, there is not much difference, making it difficult to make a judgment.

[0111] Therefore, in a preferred embodiment, an emphasis process is performed as shown in A-6 of Fig. 6 to expand the range of possible evaluation values. In this emphasis process, y = ax n Here, a is an integer of 1 or more. In a broad sense, n can be an integer of 2 or more, but if the value of y becomes too large, it becomes difficult to handle the numerical value, so the value of n is preferably about 2 to 3. In A-6 of FIG. 6, a=1, n=2, and y=x 2In other words, the sum of the threshold values ​​of the same pixel in each dither matrix is ​​squared and the resulting value is enlarged.

[0112] Then, as shown in A-7 of Fig. 6, the enlarged values ​​(highlighted values) for each pixel are summed for all pixels to obtain an evaluation value for the dither matrix group. In the example of A-7, the evaluation value is "66." The process of obtaining such an evaluation value is performed by the evaluation value calculation unit 25 of Fig. 1(B).

[0113] As explained above, in step S2-8, the average value μ and standard deviation σ of the evaluation values ​​are calculated based on the data of the evaluation values ​​that have already been acquired. This process is performed by the statistical value calculation unit 26 in FIG. 1(B).

[0114] In step S2-9, the evaluation value with the smallest numerical value is identified from among the evaluation values ​​already obtained. However, since this is a relative identification based on a comparison of numerical values, it cannot be said to be sufficiently reliable. Therefore, in step S2-10, an objective index obtained based on statistical values ​​is used to determine whether the evaluation value identified in step S2-9 can be said to be the smallest even from an objective perspective.

[0115] Specifically, it is determined whether the condition "i>q (q is, for example, 1000) and the smallest evaluation value in the i-th iteration <μ-kσ (k is, for example, 3)" is satisfied. Note that the above inequality sign ">" may be replaced with "≧". Also, the above inequality sign "<" may be replaced with "≦".

[0116] The value of variable i indicates the number of times the evaluation value has been calculated, so ensuring a sufficient number of calculations ensures the reliability of the statistical values. Then, it is determined whether the smallest evaluation value (the evaluation value identified in step S2-9) among the evaluation values ​​calculated i times satisfies the condition μ-kσ (where k is, for example, 3).

[0117] Now, reference is made to Figure 7. For example, taking a normal distribution (Gaussian distribution) as an example, values ​​that fall within a region Z1 that is -3σ or more away from the mean value μ can be determined to be sufficiently small in view of the overall distribution.

[0118] Continuing the explanation, returning to Fig. 6, if the result of step S2-10 is N, then in step S2-11 i is incremented by 1 and the same processing is repeated. If the result of step S2-10 is Y, then the process proceeds to step S2-12.

[0119] In step S2-12, a dither matrix group having an evaluation value that satisfies the condition is adopted (selected). This process is performed by the dither matrix group selection unit 27 in FIG.

[0120] In this way, in this embodiment, the evaluation value can be obtained by repeating general calculations (addition, multiplication, etc.) by focusing only on the threshold values ​​of the reference dither matrix that have small numerical values, and no particularly complex algorithm is required.

[0121] For example, as in Patent Document 2 mentioned above, there is no need for complex data processing, such as calculating an evaluation value using reinforcement learning, a form of machine learning, based on the graininess of a binary image when multiple inks are used and the incidence rate of ink dots.

[0122] Therefore, a dither matrix generating method is realized that can generate a plurality of dither matrices that can effectively suppress the graininess of an image without using a complex algorithm or the like.

[0123] Furthermore, objective evaluation using evaluation values ​​makes it possible to select a group of dither matrices that can form images with less noticeable graininess, which is also effective in automating dither matrix generation without relying on the experience of a skilled worker.

[0124] Next, reference is made to Fig. 8. Fig. 8 is a flowchart showing an example of a procedure for ranking dither matrices (an example of a procedure for determining the number of rankings with high priority). The processing in Fig. 8 is performed by the dither matrix ranking processing unit 28 in Fig. 1(A).

[0125] In step S3-1, the number p of the highest rankings that you want to prioritize is specified. For the c dither matrices included in the dither matrix group adopted in step S2-12 of FIG. 3, the number p of the highest rankings that you want to prioritize is specified. In other words, you specify how many rankings from 1 you want to prioritize. This makes it possible to prioritize consideration of the dispersibility of strong inks (such as black ink), for example.

[0126] Next, in step S3-2, it is determined whether p≧1. If the answer is N, that is, p=0, then the process proceeds to step S3-12 in FIG. 11 via branch point A (this will be described later). If the answer is Y, the process proceeds to step S3-3.

[0127] In steps S3-3 and S3-4, a process for determining a combination with a high priority is performed. In step S3-3, the evaluation value explained in step S2-7 of FIG. 3 is calculated for all combinations of p dither matrices extracted from the c dither matrix group F.

[0128] In step S3-4, the combination of p dither matrices with the smallest evaluation value is identified. This dither matrix set becomes the combination with the highest priority. In other words, the combination of p dither matrices that determines the order of priority is extracted. Because the extracted set of p dither matrices has a small evaluation value, it is a combination with a low probability of ink being deposited in the same position.

[0129] In the case of p=1, there is no difference in the evaluation value, so one image is selected at random. For example, the image that was generated earliest in the order of shift amounts in step S2-4 of FIG. 3 may be selected. Alternatively, the reference dither matrix may be selected.

[0130] Next, a dither matrix prioritization process is performed (steps S3-5 to S3-10). In step S3-5, j=p is set.

[0131] In step S3-6, it is determined whether j≧3. If N, the process proceeds to step S3-11 in FIG. 11 via branch point B (this will be described later). If Y, the process proceeds to step S3-7.

[0132] In step S3-7, (j-1) dither matrices are extracted from the j dither matrices, and the evaluation values ​​described in step S2-7 of FIG. 3 are calculated for all combinations.

[0133] In step S3-8, the combination of (j-1) dither matrices with the smallest evaluation value in step S3-7 is identified. These (j-1) dither matrices form a set of dither matrices with priority rankings 1 to (j-1).

[0134] In step S3-9, the dither matrix that was not included in the combination identified in step S3-8 is ranked j. Since the 1st to (j-1)th ranked sets of j dither matrices have been determined, the remaining dither matrix is ​​the lowest (jth ranked) of the j dither matrices. In this way, the jth ranked dither matrix is ​​determined.

[0135] In step S3-10, j is decremented by 1. Then, the process returns to step S3-6 and the same process continues. This process continues until j=2. If two dithers are not ranked, one is randomly ranked first or second. If one dither is not ranked, it is ranked first (if p=1). When j=2, the process moves to step S3-11 in FIG. 11 via branch point B.

[0136] In this way, in the dither matrix ranking process (steps S3-5 to S3-10), evaluation values ​​are calculated for all combinations of dither matrices that are one number smaller than the predetermined number from among a predetermined number of dither matrices, the combination of dither matrices with the smallest evaluation value is identified, and one dither matrix that is not included in the identified set is made the lowest ranked, and if necessary, the predetermined number is updated in the direction of decreasing that number by one and the same process is repeated, and the lowest ranked dither matrix is ​​determined for each process, thereby performing the process of identifying the priority order of the dither matrices.

[0137] By this process, the third to pth ranked dither matrices, excluding the first and second ranked ones, are determined from the p dither matrices.

[0138] Hereinafter, a specific example of the process of determining a combination with a high priority and the process of ranking dither matrices will be described with reference to FIGS.

[0139] 9 is a diagram showing a specific example of the process for determining a combination of dither matrices with high priority The process described in FIG. 9 corresponds to step S3-3 and step S4-4 in FIG.

[0140] 9, c=8 and p=4 are set, and therefore eight dither matrices (id1 to id8) are prepared. From these, a set of four dither matrices with the highest priority is determined.

[0141] In step S3-3-2, evaluation values ​​E1 to E70 are calculated for each of all combinations (70 sets) of four dither matrices selected from the eight dither matrices.

[0142] In step S3-4-1, the smallest evaluation value is identified as E1. In step S3-4-2, the set of dither matrices id1 to id4 corresponding to the identified evaluation value E1 is extracted (identified) as a combination with a high priority.

[0143] In this embodiment, the smaller the threshold value of the dither matrix, the lower the density is set, so the minimum evaluation value is calculated, but if the larger the threshold value of the dither matrix, the lower the density is set, the maximum evaluation value can be calculated, i.e., the evaluation value that causes the least deterioration in granularity can be calculated.

[0144] Next, reference will be made to Fig. 10. Fig. 10 is a diagram specifically showing the procedure for the dither matrix ranking process.

[0145] In step S3-7 of Fig. 10, j is set to 4. Therefore, evaluation values ​​E101 to E104 are calculated for all combinations of selecting three cards from four cards.

[0146] In step S3-8, the smallest evaluation value E102 is identified. In step S3-8, the dither matrices id1, id2, and id4 included in the set of dither matrices corresponding to the identified evaluation value E102 are identified as dither matrices with first to third priorities. Therefore, the dither matrix id3, which is not included in that set, is identified as fourth (jth) priority.

[0147] Next, the value of j is decremented by 1 to set j to 3. As a result of performing the same process, the set of dither matrices with the smallest evaluation value is determined. This identifies the dither matrices id1 and id2 included in this set as either the first or second ranked dither matrices.

[0148] Therefore, the dither matrix id4 that was not included in that group becomes the third-place (j-th place) dither matrix. In this way, the four dither matrices in the group with the highest priority extracted in the process of FIG. 9 are ranked third and fourth. The first- and second-place dither matrices are identified in step S3-11 of FIG. 11. This point will be described later.

[0149] Next, reference is made to Fig. 11. Fig. 11 is a flowchart showing an example of the procedure for ranking dither matrices (the procedure following Fig. 8).

[0150] In step S3-11, any of the p dither matrices that have not yet been ranked are ranked. The ranking may be determined randomly or in accordance with the order in which the shift amounts are generated.

[0151] If the top two dither matrices have no ranking (if the first and second places have not been decided), for example, one of them is randomly ranked first and the other is ranked second. Also, if one dither matrix has no ranking (if p=1), that dither matrix is ​​ranked first.

[0152] In step S3-12, the value of the variable k is set to p.

[0153] In step S3-13, for all combinations in which one dither matrix is ​​extracted from the (ck) dither matrices whose order has not been determined, the evaluation value of step S2-8 is calculated when the k dither matrices whose order has been determined and the extracted dither matrix are combined.

[0154] 8, if it is determined that p=0 and then step S3-13 is performed (in the case of processing via branch point A), then p=0 and no difference occurs in the evaluation value. In this case, for example, one dither matrix is ​​selected at random. Note that the earliest dither matrix in the order of shift amount generation may be selected, or a reference dither matrix may be selected.

[0155] In step S3-14, k=k+1 is set.

[0156] In step S3-15, the set of dither matrices with the smallest evaluation value calculated in step S3-13 is identified, and the rank of the unranked dither matrix included in that set is set to k-th place.

[0157] In step S3-16, it is determined whether k has become (c-1) (in other words, whether there is one unranked dither matrix among the c dither matrices). If it is N, the process returns to step S3-13. The same process is repeated until the (p+1)th to (c-1)th places are determined. If it is Y, the process proceeds to step S3-17.

[0158] In step S3-17, the dither matrix that has not been ranked to the end (in other words, the last dither matrix) is set to the lowest rank (c rank).

[0159] In this way, a set of dither matrices is created by adding one dither matrix selected from a plurality of dither matrices whose priority is not specified to two or more dither matrices whose priority is specified, an evaluation value is calculated for the set of dither matrices, and similar processing is repeated to identify the set with the smallest evaluation value from all sets of dither matrices obtained by updating the one dither matrix to be added, and the added dither matrix (the one whose priority is not determined) included in the identified set of dither matrices is assigned the next rank of the priority already specified, and similar processing is repeated as necessary to perform the processing of identifying the priority of the dither matrices. This completes the ranking of all c (e.g., 8) dither matrices.

[0160] When input image data is screened using c (for example, 8) dither matrices, the threshold values ​​of the dither matrices are converted inversely to values ​​before threshold conversion before screening is performed.

[0161] Currently, the number of ink colors used in inkjet printers is typically four or eight, but this number is likely to increase in the future. This embodiment can flexibly and easily accommodate an increase in the number of ink colors. In other words, by repeatedly performing a process of creating a set of dither matrices, calculating the evaluation value of that set, and finding the set with the smallest value, it is possible to easily accommodate an increase in the number of ink colors used.

[0162] Next, reference will be made to Fig. 12. Fig. 12 is a diagram showing a specific example of the dither matrix ranking process shown in Fig. 11. Fig. 12 is a continuation of the process shown in Fig. 10, which has been previously described.

[0163] In step S3-11 of FIG. 12, of the four dither matrices (id1, id2, id4, id3), dither matrices id1 and id2 are not ranked. Here, they are selected randomly, with id1 ranked first and id2 ranked second. As a result, of the eight (p) dither matrices, the first to fourth places are determined.

[0164] In step S3-13, one of the four dither matrices whose order has not been determined is added to the set of four dither matrices whose order has been determined, to create a set of five dither matrices. This creates four sets. Then, evaluation values ​​E201 to E204 are calculated for each of these four sets.

[0165] In step S3-15 of FIG. 12, k=5 is set. Here, the smallest evaluation value E203 is identified. As a result, the set of dither matrices (id1, id2, id3, id4, id7) with the smallest evaluation value is determined. Then, the unranked dither matrix id7 is determined to be in fifth place (kth place).

[0166] Thereafter, the value of k is incremented by 1 and the same process is carried out. When k=6, dither matrix id5 is determined as the 6th place (kth place). When k=7, dither matrix id4 is determined as the 7th place (kth place). With this, the dither matrices from 1st place to 7th place have been determined.

[0167] c=8 is set, and there is one remaining. Therefore, the one dither matrix id6 that has not been ranked until the end is ranked 8th (cth place). This means that all 8 dither matrices have been ranked.

[0168] Next, reference is made to Fig. 13. Fig. 13 is a flowchart showing an example of the procedure for ink ranking processing and processing for associating ink with dither matrices. The processing in Fig. 13 is performed by the ink and dither matrix association processing unit 30 in Fig. 1(A).

[0169] In step S10, the inks are ranked by comprehensively taking into consideration factors such as how easily they affect graininess, how easily a single dot of ink is perceived (strong), or how easily they are used to express pale colors.

[0170] Below is an example of how inks are ranked: 1st place: Lk (Light Black) 2nd place: K (Black) 3rd place: Lm (light magenta) 4th place: M (Magenta) 5th place: Re (Red) 6th place: Or (Orange) 7th place: Lc (Light Cyan) 8th place: C (Cyan) 9th place: Gr (Gray) 10th place: Y (Yellow) 11th place: Wh (White)

[0171] In step S11, if there is any ink that is not being used, that ink is removed and the other inks are moved up in rank.

[0172] For example, in the case of CMYK printing, the first place is K, the second place is M, the third place is C, and the fourth place is Y.

[0173] For example, in the case of CMYKOrGrReLk printing, the first place is Lk, the second place is K, the third place is M, the fourth place is Re, the fifth place is Or, the sixth place is C, the seventh place is Gr, and the eighth place is Y.

[0174] In step S12, inks with the same ranking are associated with dither matrices.

[0175] This determines which dither matrix to use for each of all the inks used.

[0176] Even if the number of ink colors used increases, the ink ranking and the correspondence between the inks and the dither matrix can be easily performed according to the above procedure. In other words, this embodiment is highly versatile. Therefore, it will be possible to flexibly and easily accommodate future increases in ink colors.

[0177] As explained above, by focusing on only the small threshold values ​​among the threshold values ​​of the reference dither matrix, an evaluation value can be found by repeating general operations (addition, multiplication, etc.), and no particularly complicated algorithm is required.

[0178] For example, as in Patent Document 2 mentioned above, there is no need for complex data processing, such as calculating an evaluation value using reinforcement learning, a form of machine learning, based on the graininess of a binary image when multiple inks are used and the incidence rate of ink dots.

[0179] Therefore, for example, it is possible to realize a dither matrix generating method that can generate a plurality of dither matrices that can effectively suppress the graininess of an image without using a complex algorithm or the like.

[0180] Furthermore, objective evaluation using evaluation values ​​makes it possible to select a group of dither matrices that can form images with less noticeable graininess, which is also effective in automating dither matrix generation without relying on the experience of a skilled worker.

[0181] Furthermore, the present invention is highly versatile, and even if the number of ink colors used increases, it is easy to associate them with the dither matrix. Therefore, it is possible to flexibly and easily accommodate future increases in ink colors.

[0182] Although the present invention has been described above using the embodiments, the present invention is not limited to these embodiments and various modifications and applications are possible.

[0183] The present invention is not limited to the above-described exemplary embodiments, and those skilled in the art will be able to easily modify the above-described exemplary embodiments to the extent that they fall within the scope of the claims. [Explanation of symbols]

[0184] 10...Reference dither matrix generation unit, 12...Dither matrix generation unit, 20...Dither matrix group generation unit, 21...Threshold conversion unit, 22...Shift amount acquisition unit, 23...Shift processing unit, 24...Dither matrix group creation unit, 25...Dither matrix group evaluation value calculation unit, 26...Statistical value calculation unit, 27...Dither matrix group selection unit, 28...Dither matrix ranking processing unit, 30...Ink and dither matrix correspondence processing unit, 40...Record Memory unit (memory), 100...dither matrix generator (image processing device), 102...data processing unit, 110...control unit (processor), 120...memory unit (memory), 200...inkjet printer (printing device), 210...input image processing device, 212...screening processing unit using dither matrix, 214...memory unit (memory), 220...printing engine, 222...head controller, 224...printing head, PRG...program.

Claims

1. A dither matrix generating method for generating a plurality of dither matrices by shifting threshold values ​​of a reference dither matrix in at least one direction selected from a first direction and a second direction orthogonal to the first direction, comprising: A step of acquiring shift amounts at least equal in number to the number of the plurality of dither matrices to be generated; a step of selecting, from the acquired shift amounts, a set of shift amounts equal in number to the number of the plurality of dither matrices to be generated, performing shift processing based on each selected shift amount to create a dither matrix having the same number of shifted dither matrices as the number of selected shift amounts, and adding the reference dither matrix to the created dither matrices to create a dither matrix group, thereafter updating the set of shift amounts, and updating the dither matrix group each time the shift amount is updated, thereby acquiring a required number of dither matrix groups; a step of calculating an evaluation value indicating the degree of deterioration of granularity when an output image is obtained by a dither method for each of the acquired dither matrix groups, selecting a dither matrix group with the smallest degree of deterioration of granularity, and generating the plurality of dither matrices using the selected dither matrix group; and A dither matrix generating method, which calculates the evaluation value using the numerical value of a threshold value of each dither matrix included in the acquired dither matrix group.

2. calculating the evaluation value using a threshold value that indicates a density lower than a predetermined density among the threshold values ​​of the reference dither matrix; The method for generating a dither matrix according to claim 1 .

3. A method for generating a dither matrix, which generates a plurality of dither matrices by shifting a threshold value of a reference dither matrix in at least one direction selected from a first direction and a second direction perpendicular to the first direction, comprising: A step of acquiring shift amounts at least equal in number to the number of the plurality of dither matrices to be generated; a step of selecting, from the acquired shift amounts, a set of shift amounts equal in number to the number of the plurality of dither matrices to be generated, performing shift processing based on each selected shift amount to create a dither matrix having the same number of shifted dither matrices as the number of selected shift amounts, and adding the reference dither matrix to the created dither matrices to create a dither matrix group, thereafter updating the set of shift amounts, and updating the dither matrix group each time the shift amount is updated, thereby acquiring a required number of dither matrix groups; a step of calculating an evaluation value indicating the degree of deterioration of granularity when an output image is obtained by a dither method for each of the acquired dither matrix groups, selecting a dither matrix group with the smallest degree of deterioration of granularity, and generating the plurality of dither matrices using the selected dither matrix group; and In the step of acquiring the shift amount, The similarity between the candidate shift amount and the shift amount that has already been adopted is determined, and the shift amount with high similarity is eliminated and the shift amount with low similarity is adopted. The dither matrix generation method.

4. A method for generating a dither matrix, which generates a plurality of dither matrices by shifting a threshold value of a reference dither matrix in at least one direction selected from a first direction and a second direction perpendicular to the first direction, comprising: A step of acquiring shift amounts at least equal in number to the number of the plurality of dither matrices to be generated; a step of selecting, from the acquired shift amounts, a set of shift amounts equal in number to the number of the plurality of dither matrices to be generated, performing shift processing based on each selected shift amount to create a dither matrix having the same number of shifted dither matrices as the number of selected shift amounts, and adding the reference dither matrix to the created dither matrices to create a dither matrix group, thereafter updating the set of shift amounts, and updating the dither matrix group each time the shift amount is updated, thereby acquiring a required number of dither matrix groups; a step of calculating an evaluation value indicating the degree of deterioration of granularity when an output image is obtained by a dither method for each of the acquired dither matrix groups, selecting a dither matrix group with the smallest degree of deterioration of granularity, and generating the plurality of dither matrices using the selected dither matrix group; and The evaluation value is summing the threshold values ​​of the same pixel in each dither matrix constituting the dither matrix group to obtain a sum value for each pixel; The sum of the pixels is set as a variable x, and the value of y is calculated using a function expressed as y=axn (a is an integer of 1 or more, and n is an integer of 2 or more). The y values ​​are summed for all pixels, and the sum is used as the evaluation value of the dither matrix group. It is calculated by The dither matrix generation method.

5. A method for generating a dither matrix, which generates a plurality of dither matrices by shifting a threshold value of a reference dither matrix in at least one direction selected from a first direction and a second direction perpendicular to the first direction, comprising: A step of acquiring shift amounts at least equal in number to the number of the plurality of dither matrices to be generated; a step of selecting, from the acquired shift amounts, a set of shift amounts equal in number to the number of the plurality of dither matrices to be generated, performing shift processing based on each selected shift amount to create a dither matrix having the same number of shifted dither matrices as the number of selected shift amounts, and adding the reference dither matrix to the created dither matrices to create a dither matrix group, thereafter updating the set of shift amounts, and updating the dither matrix group each time the shift amount is updated, thereby acquiring a required number of dither matrix groups; a step of calculating an evaluation value indicating the degree of deterioration of granularity when an output image is obtained by a dither method for each of the acquired dither matrix groups, selecting a dither matrix group with the smallest degree of deterioration of granularity, and generating the plurality of dither matrices using the selected dither matrix group; and When selecting a dither matrix group that minimizes the degree of deterioration of granularity, Among the evaluation values ​​of the dither matrix group, an evaluation value that causes the least deterioration in granularity is identified, and moreover, The number of calculated evaluation values ​​is equal to or exceeds a predetermined number, and, The specified evaluation value is equal to or less than the level at which the granularity is deteriorated, which is indicated by an index value obtained by performing statistical processing on the calculated evaluation value. Provided that, If the condition is satisfied, a dither matrix group corresponding to the specified evaluation value is selected; If the condition is not met, the dither matrix group is updated and the calculation of the evaluation value is repeated until the condition is met. The dither matrix generation method.

6. further comprising a step of assigning priorities to the generated dither matrices. The method for generating a dither matrix according to any one of claims 1 to 5.

7. A method for generating a dither matrix, which generates a plurality of dither matrices by shifting a threshold value of a reference dither matrix in at least one direction selected from a first direction and a second direction perpendicular to the first direction, comprising: A step of acquiring shift amounts at least equal in number to the number of the plurality of dither matrices to be generated; a step of selecting, from the acquired shift amounts, a set of shift amounts equal in number to the number of the plurality of dither matrices to be generated, performing shift processing based on each selected shift amount to create a dither matrix having the same number of shifted dither matrices as the number of selected shift amounts, and adding the reference dither matrix to the created dither matrices to create a dither matrix group, thereafter updating the set of shift amounts, and updating the dither matrix group each time the shift amount is updated, thereby acquiring a required number of dither matrix groups; a step of calculating an evaluation value indicating the degree of deterioration of granularity when an output image is obtained by a dither method for each of the acquired dither matrix groups, selecting a dither matrix group with the smallest degree of deterioration of granularity, and generating the plurality of dither matrices using the selected dither matrix group; a step of assigning priorities to the generated dither matrices; and In the step of assigning priority, a process of calculating the evaluation values ​​for all combinations of dither matrices, which are selected from a predetermined number of dither matrices and have a number smaller than the predetermined number by one, identifying a combination of dither matrices that results in an evaluation value that causes the least deterioration in granularity, placing one dither matrix not included in the identified combination at the lowest rank, and repeating the same process by updating the predetermined number in a direction that decreases the number by one as necessary, and determining the lowest rank for each process, thereby identifying the priority of the dither matrices; and, a process of creating a set of dither matrices by adding one dither matrix selected from a plurality of dither matrices for which the priority is not specified to two or more dither matrices for which the priority is specified, calculating the evaluation value for the set of dither matrices, and subsequently repeating the same process to identify a set with the evaluation value that causes the least deterioration in granularity from among all sets of dither matrices obtained by updating the one dither matrix to be added, assigning the one added dither matrix included in the identified set of dither matrices the next rank in the priority order that has already been specified, and repeating the same process as necessary to identify the priority of the dither matrices; At least one of the following is performed: The dither matrix generation method.

8. A dither matrix generation method for generating multiple dither matrices by shifting a threshold value of a reference dither matrix in at least one direction selected from a first direction and a second direction perpendicular to the first direction, comprising: A step of acquiring shift amounts at least equal in number to the number of the plurality of dither matrices to be generated; a step of selecting, from the acquired shift amounts, a set of shift amounts equal in number to the number of the plurality of dither matrices to be generated, performing shift processing based on each selected shift amount to create a dither matrix having the same number of shifted dither matrices as the number of selected shift amounts, and adding the reference dither matrix to the created dither matrices to create a dither matrix group, thereafter updating the set of shift amounts, and updating the dither matrix group each time the shift amount is updated, thereby acquiring a required number of dither matrix groups; a step of calculating an evaluation value indicating the degree of deterioration of granularity when an output image is obtained by a dither method for each of the acquired dither matrix groups, selecting a dither matrix group with the smallest degree of deterioration of granularity, and generating the plurality of dither matrices using the selected dither matrix group; a step of assigning priorities to the generated dither matrices; a step of assigning a priority to each of the plurality of inks in advance and associating the inks with dither matrices having the same priority; The dither matrix generation method.

9. Inks containing black colorants are assigned to the dither matrix with the highest priority. The method for generating a dither matrix according to claim 8.

10. A dither matrix generating device that generates a plurality of dither matrices by shifting threshold values ​​of a reference dither matrix in at least one direction selected from a first direction and a second direction orthogonal to the first direction, comprising: a shift amount acquisition unit that determines shift amounts at least equal to the number of the plurality of dither matrices to be generated; a dither matrix group creating unit that creates a required number of dither matrix groups by repeatedly selecting, from the acquired shift amounts, sets of shift amounts equal in number to the number of the plurality of dither matrices to be created, performing shift processing based on each selected shift amount, creating shifted dither matrices equal in number to the selected shift amounts, and adding the reference dither matrix to the created dither matrices to create a dither matrix group, thereafter updating the sets of shift amounts, and updating the dither matrix group each time the shift amount is updated; a dither matrix generating unit that calculates an evaluation value indicating the degree of deterioration of granularity when an output image is obtained by a dither method for each of the created dither matrix groups, selects a dither matrix group that causes the smallest degree of deterioration of granularity, and generates the plurality of dither matrices using the selected dither matrix group; and The dither matrix generating unit calculates the evaluation value using a threshold value of each dither matrix included in the created dither matrix group.

11. A program that causes a computer to function as the shift amount acquisition unit, the dither matrix group creation unit, and the dither matrix generation unit according to claim 10.

Images