Threshold matrix generating method, image data generating method, program, threshold matrix, and image data generating device

Abstract

To improve graininess in a highlight area and efficiently change a density in a tone change on a highlight side without loss of reproduction in a shadow part of a print image.SOLUTION: In a threshold matrix generation method, after a plurality of periodic areas 83 are set in a matrix space 80 and elements 81 existing every other one in a row direction and a column direction in each periodic area is set as target elements 81a, a step is performed for assigning threshold values ranging from one on the most highlight side to a predetermined switching threshold value sequentially to target elements in the plurality of periodic areas. Assuming that elements to which threshold values are assigned are regarded as determined elements 82, in the step, under a condition that the number of determined elements is almost same in the plurality of periodic areas and each determined element is positioned in the vicinity of any one determined element in each periodic area, a target element whose distance to all determined elements is largest is specified within the plurality of periodic areas and a threshold value is assigned to the target element.SELECTED DRAWING: Figure 12

Description

[Technical Field]

[0001] The present invention relates to a technique for generating a threshold matrix and a technique for generating image data using a threshold matrix. [Background technology]

[0002] Conventionally, a multi-tone original image is halftone-doted and the halftone image is printed on a substrate such as paper (i.e., a printed image is formed). AM (Amplitude Modulated) screens, FM (Frequency Modulated) screens, etc. are used to halftone-dot the original image.

[0003] Patent Document 1 discloses a method for generating a threshold matrix for N-times speed. In this method, in a matrix space, drawing elements corresponding to drawing positions during N-times speed drawing are set every (N-1)th drawing element. Furthermore, a plurality of first partial areas are set, each containing a plurality of drawing elements and distributed approximately uniformly in the matrix space. Appearance numbers are assigned to two or more drawing elements in each first partial area, and then appearance numbers are assigned to the remaining drawing elements. Then, a threshold matrix for N-times speed is obtained by determining the threshold of each drawing element according to the appearance numbers. Patent Document 2 describes that by actively agglomerating dots in the low- to medium-density gradation range, robustness against streaks, which are characteristic of single-pass printing, is improved. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2015-12370 [Patent Document 2] Japanese Patent Application Laid-Open No. 2014-150510 Summary of the Invention [Problem to be solved by the invention]

[0005] However, FM screens worsen graininess in highlight areas of printed images compared to AM screens. Meanwhile, AM screens create larger overlapping dot areas in highlight areas of printed images than FM screens, making it difficult to efficiently change density (i.e., improve dot gain efficiency) when changing gradation in the highlight area. Furthermore, because AM screens have a constant period, they are inferior to FM screens in some respects, such as the occurrence of broken lines in the reproduction of fine details.

[0006] The present invention has been made in consideration of the above-mentioned problems, and aims to improve the graininess in highlight regions without impairing the reproducibility of shadow areas in printed images, and to efficiently change the density when changing the gradation on the highlight side. [Means for solving the problem]

[0007] The invention of claim 1 is a threshold matrix generation method for generating a threshold matrix to be compared with an original image when the original image having multiple tones is halftoned, the method comprising: a) a step of preparing a matrix space which is a set of elements arranged in row and column directions; b) a step of setting a plurality of periodic regions which are periodically arranged and uniformly distributed in the matrix space, each containing a plurality of elements; c) a step of setting every other element in the row and column directions in at least each periodic region as a target element; d) a step of assigning thresholds ranging from a threshold on the most highlight side to a predetermined switching threshold to the target elements included in the plurality of periodic regions in order; and e) a step of assigning remaining thresholds to the remaining elements to which no threshold has been assigned in order after the step d), wherein elements to which a threshold has been assigned are regarded as determined elements, and when assigning each threshold in the step d), the number of determined elements in the plurality of periodic regions is determined based on the number of determined elements. The difference between the maximum and minimum values ​​of is 1 or 0 and each determined element in each periodic region is one of the determined elements adjacent to, or adjacent to the 8-neighboring elements of any determined elementUnder certain conditions, while taking into account the repeated application of the threshold matrix during halftoning of the original image, the target element having the greatest distance to all determined elements is identified within the multiple periodic regions and assigned the respective threshold values.

[0008] The invention of claim 2 is the threshold matrix generation method of claim 1, wherein in the step d), a target element adjacent to any determined element is identified and a threshold is assigned.

[0009] The invention described in claim 3 is the threshold matrix generation method described in claim 1 or 2, wherein when assigning each threshold in step e), an element having the maximum distance from all determined elements is identified and assigned each threshold, taking into consideration the repeated application of the threshold matrix when halftoning the original image.

[0010] The invention described in claim 4 is the threshold matrix generation method described in claim 3, wherein in step c), every other element in the row direction and the column direction in the matrix space is set as a target element, and step e) comprises: e1) a step of assigning thresholds from the threshold next to the switching threshold to another switching threshold to the target elements among the remaining elements, in order; and e2) after step e1), a step of assigning thresholds from the threshold next to the other switching threshold to the threshold furthest from the shadow side to the remaining elements to which no threshold has been assigned.

[0011] The invention described in claim 5 is a threshold matrix generation method described in any one of claims 1 to 4, wherein a threshold matrix for a first color component and a threshold matrix for a second color component are generated by repeating steps a) to e), and the position of the target element to which the most highlight-side threshold is first assigned in step d) when the threshold matrix for the first color component is generated differs from that when the threshold matrix for the second color component is generated.

[0012] The invention described in claim 6 is a threshold matrix generation method described in any one of claims 1 to 5, wherein by repeating steps a) to e), a threshold matrix for a first color component and a threshold matrix for a second color component are generated, and the arrangement, shape, or size of the plurality of periodic regions when the threshold matrix for the first color component is generated differs from that when the threshold matrix for the second color component is generated.

[0013] The invention described in claim 7 is an image data generation method for generating image data, comprising the steps of: preparing a threshold matrix generated by the threshold matrix generation method described in any one of claims 1 to 6; and generating halftone image data by halftoning a multi-tone original image by comparing the original image with the threshold matrix.

[0014] The invention of claim 8 is a program for causing a computer to generate a threshold matrix to be compared with a multi-tone original image when the original image is halftone-coded, wherein execution of the program by a computer causes the computer to execute the following steps: a) preparing a matrix space which is a set of elements arranged in row and column directions; b) setting a plurality of periodic regions in the matrix space which are periodically arranged and uniformly distributed, each including a plurality of elements; c) setting, in at least each periodic region, every other element in the row and column directions as a target element; d) assigning thresholds ranging from the most highlight-side threshold to a predetermined switching threshold to the target elements included in the plurality of periodic regions in order; and e) after step d), assigning the remaining thresholds in order to the remaining elements to which no threshold has been assigned, wherein elements to which a threshold has been assigned are considered as determined elements, and when assigning each threshold in step d), the number of determined elements in the plurality of periodic regions is determined. The difference between the maximum and minimum values ​​of is 1 or 0 and each determined element in each periodic region is one of the determined elements adjacent to, or adjacent to the 8-neighboring elements of any determined elementUnder certain conditions, while taking into account the repeated application of the threshold matrix during halftoning of the original image, the target element having the greatest distance to all determined elements is identified within the multiple periodic regions and assigned the respective threshold values.

[0015] The invention described in claim 9 is a threshold matrix that is compared with an original image when halftoning a multi-tone original image, and when a plurality of halftone images ranging from the most highlight-side gradation value to a predetermined switching gradation value are generated, dots are formed only at target pixels that exist in every other pixel in the row and column directions in a plurality of periodic regions that are periodically arranged and uniformly distributed in the plurality of halftone images, each of which includes a plurality of pixels, and the positions at which dots are added from a halftone image of one gradation value to a halftone image of the next gradation value in the plurality of halftone images are determined by the number of dots in the plurality of periodic regions in each halftone image. The difference between the maximum and minimum values ​​of is 1 or 0 and each dot in each periodic region is one of the dots or adjacent to the 8-neighboring pixels of any dot This is the target pixel that has the maximum distance to all existing dots under the given conditions.

[0016] The invention described in claim 10 is an image data generating device that generates image data, and includes a matrix storage unit that stores the threshold matrix described in claim 9, and an image data generating unit that generates halftone image data by halftoning a multi-tone original image by comparing the original image with the threshold matrix. [Effects of the Invention]

[0017] According to the present invention, it is possible to improve graininess in highlight regions without impairing the reproducibility of shadow areas in a printed image, and to efficiently change density in gradation changes on the highlight side. [Brief explanation of the drawings]

[0018] [Figure 1] FIG. 1 illustrates a configuration of a printing device. [Figure 2] FIG. 1 illustrates the configuration of a computer. [Figure 3] FIG. 2 is a block diagram illustrating the functional configuration of the printing apparatus. [Figure 4] FIG. 10 is a diagram showing the flow of processing for printing an image. [Figure 5] FIG. 2 is a diagram illustrating a threshold matrix and an original image. [Figure 6] FIG. 10 is a diagram showing a processing flow for generating a threshold matrix. [Figure 7] FIG. 10 is a diagram showing the flow of highlight-side threshold allocation processing. [Figure 8] FIG. 1 is a diagram illustrating a matrix space. [Figure 9] FIG. 1 is a diagram illustrating a matrix space. [Figure 10] FIG. 10 is a diagram showing how the number of determined elements increases. [Figure 11] FIG. 10 illustrates determined element clusters in multiple periodic regions. [Figure 12] FIG. 1 is a diagram illustrating a matrix space. [Figure 13] FIG. 10 is a diagram showing the flow of an intermediate threshold value allocation process. [Figure 14] FIG. 1 is a diagram illustrating a matrix space. [Figure 15] FIG. 1 is a diagram showing a halftone dot image. [Figure 16] FIG. 10 is a diagram showing the flow of a shadow-side threshold allocation process. [Figure 17] FIG. 1 is a diagram showing a halftone dot image. [Figure 18] FIG. 1 is a diagram showing a halftone dot image. [Figure 19A] FIG. 1 is a diagram showing a periodic region. [Figure 19B] FIG. 1 is a diagram showing a periodic region. [Figure 20] FIG. 1 is a diagram showing a halftone dot image of multiple color components. [Figure 21] FIG. 1 is a diagram showing a halftone dot image of multiple color components. [Figure 22] 1 is a photograph showing highlight and shadow areas of a printed image. [Figure 23] 10 is a photograph showing highlight and shadow areas of a printed image of a comparative example. [Figure 24] 1 is a photograph showing a printed image. [Figure 25] 10 is a photograph showing a printed image of a comparative example. [Figure 26A] FIG. 10 is a diagram showing another example of a periodic region. [Figure 26B] FIG. 10 is a diagram showing another example of a periodic region. [Figure 26C] FIG. 10 is a diagram showing another example of a periodic region. [Figure 26D] FIG. 10 is a diagram showing another example of a periodic region. [Figure 26E] FIG. 10 is a diagram showing another example of a periodic region. [Figure 26F] FIG. 10 is a diagram showing another example of a periodic region. [Figure 27] FIG. 1 is a diagram illustrating a matrix space. [Figure 28] FIG. 1 is a diagram illustrating a matrix space. DETAILED DESCRIPTION OF THE INVENTION

[0019] FIG. 1 is a diagram showing the configuration of a printing device 1 according to one embodiment of the present invention. The printing device 1 is a device that performs color printing using an inkjet method on a substrate such as a long sheet of printing paper or film. The substrate is not limited to paper, but will be referred to as "printing paper 9" hereinafter. In the printing device 1, images are printed in multiple areas on the printing paper 9 corresponding to multiple pages, for example. The printing device 1 may also perform monochrome printing.

[0020] 1 includes a main body 10 and a computer 5 connected to the main body 10. The main body 10 includes a discharge unit 2 that discharges minute droplets of ink toward printing paper 9, a movement mechanism 3 that moves the printing paper 9 in the (-Y) direction in FIG. 1 below the discharge unit 2, and a main body control unit 4 that is connected to the discharge unit 2 and the movement mechanism 3.

[0021] In the movement mechanism 3, multiple rollers 311, each long in the X direction (hereinafter referred to as the "width direction" since this corresponds to the width of the printing paper 9), are arranged in the Y direction. On the (+Y) side of the multiple rollers 311 is a supply unit 313 that holds a roll of printing paper 9 before printing and feeds the printing paper 9 from the roll in the (-Y) direction. On the (-Y) side of the multiple rollers 311 is a winding unit 312 that winds up and holds a section of printing paper 9 that has been printed. In the movement mechanism 3, each section of printing paper 9 moves continuously in the Y direction between the supply unit 313 and the winding unit 312. The movement mechanism 3 is supported by the side wall unit 20. In the following explanation, the Y direction is also referred to as the "movement direction." Furthermore, in the explanation of the printing device 1, when the term "printing paper 9" is used simply, it refers to the section of printing paper 9 that is in the process of moving (i.e., the section of printing paper 9 on the multiple rollers 311).

[0022] The ejection unit 2 is attached to a frame 301 that spans the width of the printing paper 9. The ejection unit 2 includes multiple head units (four in this embodiment). The multiple head units eject ink of the colors K (black), C (cyan), M (magenta), and Y (yellow), respectively, and are arranged in the Y direction. Each head unit has multiple ejection ports arranged in the width direction. For example, multiple ejection port arrays are arranged in the Y direction, with multiple ejection port arrays consisting of multiple ejection port arrays. When focusing on the width direction, one ejection port in each ejection port array is located between two adjacent ejection port arrays. This makes it possible to form multiple dots arranged in a line in the width direction on the printing paper 9 at a pitch equal to or less than the pitch of the ejection ports in each ejection port array. The ejection unit 2 may also include head units that eject ink of other colors.

[0023] In the head unit, a piezoelectric liquid ejection element, for example, is provided for each ejection port, and by driving the liquid ejection element, minute droplets of ink are ejected from each ejection port toward the printing paper 9. In reality, the multiple ejection ports are aligned across the entire width of the printing area on the printing paper 9 in the width direction, and printing of an image on the printing paper 9 is completed by the printing paper 9 passing just once below the head unit. In other words, the printing device 1 is capable of high-speed image printing using a single-pass method. The printing device 1 may also use a method other than a single-pass method. The liquid ejection element is not limited to a piezoelectric type, and for example, a thermal type, in which air bubbles are generated inside a pressure chamber by heating, may be used.

[0024] 2 is a diagram showing the configuration of the computer 5. The computer 5 has a typical computer system configuration including a CPU 501 that performs various arithmetic processing, a ROM 502 that stores basic programs, and a RAM 503 that stores various information. The computer 5 further includes an image memory 504 that stores data of original color images, a fixed disk 505 that stores information, a display 506 that displays various information, a keyboard 507a and a mouse 507b that accept input from an operator, a read / write device 508 that reads information from and writes information to a computer-readable recording medium 90 such as an optical disk, a magnetic disk, or a magneto-optical disk, and a communication unit 509 that communicates with the main body control unit 4.

[0025] In computer 5, program 900 is read in advance from recording medium 90 via read / write device 508 and stored on fixed disk 505. Program 900 may also be stored on fixed disk 505 via a network. CPU 501 executes arithmetic processing in accordance with program 900 using RAM 503 and fixed disk 505 (i.e., computer 5 executes the program), whereby computer 5 performs processing as a calculation unit that generates a threshold matrix (also called SPD (Screen Pattern Data)) used for halftoning an original image. The generated threshold matrix is ​​transferred to main body control unit 4 via communication unit 509.

[0026] FIG. 3 is a block diagram showing the functional configuration of the printing device 1. The highlight processing unit 51, intermediate processing unit 52, and shadow processing unit 53 of the calculation unit 50 in FIG. 3 are functions realized by the computer 5. The functions of the highlight processing unit 51, intermediate processing unit 52, and shadow processing unit 53 will be described in detail below. The functions of the calculation unit 50 may be realized by a dedicated electrical circuit, or a dedicated electrical circuit may be used in part. Furthermore, the calculation unit 50 may be realized by multiple computers working together, in which case the multiple computers may be located remotely from one another.

[0027] The main body control unit 4 comprises an image memory 41, multiple matrix storage units 42, a comparator 43, a movement control unit 45, and a discharge control unit 44. The image memory 41 stores data of an original color image. The multiple matrix storage units 42 are memories that store threshold matrices for multiple color components. The comparator 43 is a halftone dot circuit that compares the original image with the threshold matrix for each color component. The movement control unit 45 controls the movement mechanism 3 that moves the printing paper 9. The discharge control unit 44 controls the discharge of ink from the multiple discharge ports of the discharge unit 2 in synchronization with the movement of the printing paper 9.

[0028] Next, the process by which the printer 1 prints an image will be described with reference to Fig. 4. First, a threshold matrix to be used for printing is output from the computer 5 (calculation unit 50) to the main body control unit 4 (or may be output in advance), and is stored and prepared in the matrix storage unit 42 of Fig. 3 (step S11). The process of generating the threshold matrix will be described later. Also, a color original image is input to the main body control unit 4 from the computer 5 or an external computer, and is stored in the image memory 41.

[0029] FIG. 5 is a diagram illustrating an abstract threshold matrix 8 and an original image 70. While FIG. 5 shows only the threshold matrix 8 for one color component, the same applies to other color components. In the threshold matrix 8, multiple elements are arranged in a row direction (shown as the x direction in FIG. 5) corresponding to the width direction, and in a column direction (shown as the y direction in FIG. 5) corresponding to the movement direction. In the original image 70, multiple pixels are also arranged in a direction corresponding to the width direction (hereinafter referred to as the "row direction" as in the threshold matrix 8) and a direction corresponding to the movement direction (hereinafter referred to as the "column direction" as in the threshold matrix 8) (the same applies to halftone images described below). In the following description, the original image is represented by gradation values ​​(integer values) ranging from 0 to 255. Of course, the number of gradations of the original image may be 4096 or any other number that can be determined arbitrarily.

[0030] Next, in comparator 43, which is an image data generating unit, the original image 70 stored in image memory 41 for each color component is compared with the threshold matrix 8 stored in matrix storage unit 42. As a result, the original image 70 is halftone-coded (i.e., halftone processing is performed), and halftone image data (hereinafter simply referred to as a "halftone image") to be used for printing in printing device 1 is generated (step S12).

[0031] Here, the halftoning of the original image 70 will be explained. When halftoning the original image 70, the original image 70 is divided into a large number of regions of the same size as shown in FIG. 5, and repeat regions 71, which serve as units of halftoning, are set. Each matrix storage unit 42 has a storage area corresponding to one repeat region 71, and a threshold value is set at each address (coordinate) of this storage area to store a threshold matrix 8. Conceptually, each repeat region 71 of the original image 70 is superimposed on the threshold matrix 8 for each color component, and the gradation value of that color component of each pixel in the repeat region 71 is compared with the corresponding threshold value in the threshold matrix 8 to determine whether or not to perform drawing (form a dot of that color) at the position of that pixel on the printing paper 9.

[0032] In practice, the gradation value of one pixel in the original image 70 is read out for each color component from the image memory 41 based on an address signal from the address generator included in the comparator 43 in Fig. 3. Meanwhile, the address generator also generates an address signal indicating the position in the repeat area 71 corresponding to that pixel in the original image 70, and one threshold value in the threshold matrix 8 for each color component is identified and read out from the matrix storage unit 42. Then, the gradation value from the image memory 41 is compared with the threshold value from the matrix storage unit 42 for each color component by the comparator 43, and the gradation value of the position (address) of that pixel in the binary halftone image (output image) for each color component is determined.

[0033] Therefore, when focusing on one color component, in the multi-tone (continuous tone) original image 70 shown in Fig. 5, a tone value of "1" is assigned (i.e., a dot is placed) to positions where the tone value is greater than the corresponding threshold value in the threshold matrix 8, and a tone value of "0" is assigned (i.e., no dot is placed) to the remaining pixels. In this way, the main body control unit 4, which is an image data generating device, halftones the original image 70 using the threshold matrix 8, and halftone image data indicating the ON / OFF of ink ejection from the multiple ejection ports is generated.

[0034] 1, in parallel with the halftone dot processing (processing for generating halftone dot image data), an image is printed on printing paper 9. That is, the movement control unit 45 drives the movement mechanism 3 to start moving the printing paper 9 in the movement direction (step S13), and the ejection of ink from the multiple ejection ports included in each head unit of the ejection unit 2 is controlled by the ejection control unit 44 in synchronization with the movement of the printing paper 9 (step S14).

[0035] Here, because the halftone dot image is an image printed on the printing paper 9, the multiple pixels of the halftone dot image can be considered to be set in an array on the printing paper 9. Furthermore, the multiple pixel positions in the row direction of the halftone dot image are respectively associated with the multiple ejection ports of each head unit. In parallel with the relative movement of the ejection unit 2 with respect to the printing paper 9, the ejection control unit 44 forms a dot at each ejection position on the printing paper 9 of each ejection port when the gradation value of the halftone dot image corresponding to that ejection position on the printing paper 9 of that ejection port is "1," and does not form a dot at that ejection position when the gradation value is "0." In this way, the ejection of ink from the multiple ejection ports is controlled for each of K, C, M, and Y in accordance with the gradation value of the halftone dot image corresponding to that ejection position on the printing paper 9 of the multiple ejection ports.

[0036] In the printing device 1, halftone dot images for K, C, M, and Y are generated while the halftone dot images are printed on the printing paper 9 in parallel, and a color halftone dot image that represents the original color image is printed on the printing paper 9. When the entire halftone dot image is printed on the printing paper 9, the movement of the printing paper 9 stops, and the printing operation in the printing device 1 ends (step S15). In the following explanation, the halftone dot image printed on the printing paper 9 is referred to as the "printed image."

[0037] Next, the process of generating the threshold matrix 8 used in the printing device 1 will be described with reference to FIG. 6. As mentioned above, the threshold matrix 8 is compared with the original multi-tone image 70 when the original multi-tone image 70 is halftoned. The description of the generation of the threshold matrix 8 focuses on only one of the color components K, C, M, and Y, but the same applies to the other color components. Of course, threshold matrices 8 for color components other than K, C, M, and Y may also be generated.

[0038] In the calculation unit 50 of FIG. 3, a storage area corresponding to the above-mentioned repeat area 71 is prepared as a matrix space (step S21). The matrix space is a set of elements (matrix elements) arranged in row and column directions. Each element can store one threshold value. In the following process, a threshold value is set for each element of the matrix space, thereby generating a threshold matrix 8.

[0039] Next, the highlight processing unit 51 of the calculation unit 50 performs a process of assigning highlight-side thresholds to elements of the matrix space (hereinafter referred to as the "highlight-side threshold assignment process") (step S22). FIG. 7 is a diagram showing the flow of the highlight-side threshold assignment process. In the highlight processing unit 51, first, as shown in FIG. 8, elements 81 existing every other in the row and column directions throughout the entire matrix space 80 are set as target elements 81a (step S221). FIG. 8 shows a portion of the matrix space 80, and the target elements 81a are indicated by parallel diagonal lines (the same applies to FIGS. 9, 12, and 14 described below). If elements 81 other than the target elements 81a are called "non-target elements," the target elements 81a and non-target elements are alternately arranged in the row direction, and the target elements 81a and non-target elements are alternately arranged in the column direction. In this way, the target elements 81a and non-target elements are arranged in a checkerboard pattern.

[0040] Furthermore, as shown in FIG. 9 , multiple periodic regions 83 are set in the matrix space 80 (step S222). In FIG. 9 , each periodic region 83 is surrounded by a thick line. Each periodic region 83 includes multiple elements 81. Typically, each periodic region 83 is a region of elements 81 arranged in M ​​rows and L columns (M and L are integers equal to or greater than 2), and the multiple periodic regions 83 have the same size and shape. In the example of FIG. 9 , each periodic region 83 is a square region consisting of 5 rows and 5 columns of elements 81, and includes 13 target elements 81 a and 12 non-target elements. The multiple periodic regions 83 are periodically (regularly) arranged and uniformly distributed in the matrix space 80. Preferably, the multiple periodic regions 83 are arranged at regular intervals in both the row and column directions. In the example of FIG. 9 , the width of the gap between two adjacent periodic regions 83 in the row direction is the same as the width of the periodic regions 83 in the row direction. Similarly, the width of the gap between two periodic regions 83 adjacent to each other in the column direction is the same as the width of the periodic regions 83 in the column direction.

[0041] When the target elements 81a and periodic regions 83 are set in the matrix space 80, the highlight processing unit 51 sequentially assigns thresholds ranging from the most highlighted threshold to a predetermined first switching threshold to the target elements 81a included in the multiple periodic regions 83. In detail, first, the most highlighted threshold (initial value) is set as the current threshold, and the periodic region 83 to be targeted in the assignment of the current threshold is determined as the target periodic region 83 (step S223). Here, if the elements to which thresholds are assigned are called "determined elements," the target periodic region 83 is a periodic region 83 that has a smaller number of determined elements than the other periodic regions 83. In the initial threshold assignment, none of the periodic regions 83 include determined elements, and therefore all of the periodic regions 83 become the target periodic regions 83.

[0042] Next, a target element 81a to which the current threshold should be assigned is determined in each of the target periodic regions 83. Here, the number of determined elements in each target periodic region 83 is 0 (step S224), and none of the periodic regions 83 includes a determined element, so the target element 81a to which the threshold should be assigned is determined randomly. Then, the current threshold, i.e., the threshold most on the highlight side, is assigned to the target element 81a (step S225). Note that the assignment of a threshold to an element 81 can also be considered as the arrangement of dots for the element 81.

[0043] In this embodiment, the original image is expressed using tone values ​​ranging from 0 to 255, with 0 being the most highlight-side threshold and 254 being the most shadow-side threshold. Each integer value from 0 to 254 is assigned as a threshold to approximately the same number of elements 81 (i.e., the number of integer parts of the value obtained by dividing the number of elements in matrix space 80 by 255, or the number obtained by adding 1 to the integer part). The number of elements 81 to which each value from 0 to 254 is assigned does not necessarily have to be approximately the same; they may differ depending on factors such as the characteristics of the ink used in the printing device 1. For example, the number of elements 81 to which a highlight-side threshold is assigned may be fewer than the number of elements 81 to which a shadow-side threshold is assigned. In the following description, the number of elements 81 to which each value within the entire threshold range (here, 0 to 254) is assigned is referred to as the "set number."

[0044] In this case, since the current threshold has not yet been assigned to the set number of elements 81, the current threshold is determined as the next threshold (i.e., the next current threshold) (step S226). Next, the current threshold is compared with the first switching threshold described above. The first switching threshold is a value that is 25% or less (in this processing example, it is any value between 0 and 63, and hereinafter referred to as "a value that is 25% or less of the entire threshold range"), where the threshold on the most highlight side is 0% and the threshold on the most shadow side is 100%. The same applies hereinafter. The first switching threshold is preferably a value that is 15% or more of the entire threshold range, and more preferably a value that is 18% or more of the entire threshold range (the reason for this will be explained later).

[0045] Because the current threshold is equal to or less than the first switching threshold (step S227), the process returns to step S223 to determine the target periodic region 83. As described above, the target periodic region 83 is a periodic region 83 having a smaller number of determined elements than the other periodic regions 83, so a periodic region 83 other than the periodic region 83 including the target element 81 a to which a threshold was assigned in step S225 above is determined to be the target periodic region 83.

[0046] After confirming that the number of determined elements in the target periodic region 83 is 0 (step S224), a target element 81a to which the current threshold should be assigned is determined. Specifically, a target element 81a for which a threshold has not been determined and which has the greatest distance from all determined elements is identified within the multiple target periodic regions 83. At this time, assuming that the same matrix space 80 is also arranged in the vicinity of the matrix space 80, a target element 81a (within the target periodic region 83) which has the greatest distance from all determined elements is identified in the central matrix space 80. In this way, the identification of the target element 81a takes into account the repeated application of the threshold matrix 8 when halftoning the original image 70. To identify the target element 81a which has the greatest distance from all determined elements, for example, an equation similar to Mathematical Formula 1 in Japanese Patent Laid-Open No. 2015-12370 (the above-mentioned Patent Document 1) can be used (the same applies to steps S232 and S242 described below). Then, the current threshold value is assigned to the identified target element 81a (step S225).

[0047] The above steps S223 to S225 are repeated until the current threshold is assigned to the set number of elements 81 (steps S226, S227). Once the current threshold is assigned to the set number of elements 81, the value obtained by adding 1 to the current threshold is determined as the next threshold (i.e., the next current threshold) (step S226). Then, the above steps S223 to S225 are repeated until the current threshold is assigned to the set number of elements 81 (steps S226, S227).

[0048] In step S223 of the repetition of steps S223 to S225, when one target element 81a becomes a determined element in all periodic regions 83, there are no periodic regions 83 with a smaller number of determined elements than the other periodic regions 83, so all periodic regions 83 become target periodic regions 83. Furthermore, since the number of determined elements in the target periodic region 83 is not 0 (step S224), only target elements 81a adjacent to the determined elements (for which a threshold has not been determined), i.e., target elements 81a that are diagonally adjacent to the determined elements, become candidates for the target element 81a to which the current threshold should be assigned. Then, among the target elements 81a adjacent to any determined element in the target periodic region 83, the target element 81a that is the longest distance from all the determined elements is identified, and the current threshold is assigned to that target element 81a (step S228). In this way, when the number of determined elements in the target periodic region 83 becomes one or more, the process of step S228 is performed instead of step S225.

[0049] In the subsequent processing, when the current threshold has been assigned to a set number of elements 81, the value obtained by adding 1 to the current threshold is determined as the next threshold, and the above steps S223, S224, and S228 are repeated (steps S226 and S227). In step S223 during the repetition of the above steps S223, S224, and S228, as described above, a periodic region 83 having a smaller number of determined elements than the other periodic regions 83 is determined as the target periodic region 83. In step S228, the current threshold is assigned to the target element 81a that is the longest distance from all of the determined elements among the target elements 81a in the target periodic region 83 adjacent to any of the determined elements.

[0050] FIG. 10 is a diagram showing an example of how the number of determined elements 82 increases in one periodic region 83. In FIG. 10, the determined elements 82 (elements 81 to which a threshold value has been assigned) are painted black, i.e., dots are placed on the determined elements 82 (similarly in other diagrams showing the matrix space 80). In addition, the number of repetitions of steps S223, S224, and S228 increases from the leftmost determined element 82 toward the cluster of determined elements 82 (cluster of dots) on the rightmost side. As shown in FIG. 10, the cluster of dots becomes larger (grows) as the number of repetitions of steps S223, S224, and S228 increases. As described above, in step S228, a target element 81a to which a threshold value should be assigned is identified from target elements 81a adjacent to any determined element 82, and therefore, each determined element 82 in each periodic region 83 is adjacent to any determined element 82.

[0051] FIG. 11 is a diagram showing clusters of determined elements 82 in multiple periodic regions 83 arranged side by side, illustrating clusters of determined elements 82 at a certain stage in the highlight-side threshold allocation process. As shown in FIG. 11, the arrangement of the determined elements 82 varies in the multiple periodic regions 83. Furthermore, as described above, a periodic region 83 having a smaller number of determined elements 82 than the other periodic regions 83 is determined as the target periodic region 83, and therefore the difference between the maximum and minimum numbers of determined elements 82 included in each periodic region 83 at the same stage is 1 or 0. In other words, when assigning each threshold in the highlight-side threshold allocation process, the number of determined elements 82 in the multiple periodic regions 83 is approximately the same.

[0052] 11 shows the arrangement of dots in the area of ​​the halftone image (hereinafter also referred to as "periodic area") corresponding to each periodic area 83 when a uniform image is halftone-coded using the threshold matrix 8 at a gradation value that is one greater than the threshold value at the above stage (i.e., when a halftone image of that gradation value is generated using the threshold matrix 8). Even in that halftone image (halftone image), the arrangement of dots in the multiple periodic areas varies in various ways, giving rise to fluctuations.

[0053] When the current threshold value becomes greater than the first switching threshold value during the repetition of steps S223, S224, and S228 (Yes in step S227), the highlight-side threshold allocation process ends. Fig. 12 is a diagram showing a portion of the matrix space 80 at the end of the highlight-side threshold allocation process. As shown in Fig. 12, at the end of the highlight-side threshold allocation process, only the target elements 81a included in the periodic regions 83 are determined elements 82, and no determined elements 82 exist in regions other than the periodic regions 83. Furthermore, all periodic regions 83 include approximately the same number of determined elements 82, and the determined elements 82 in each periodic region 83 form a certain amount of clusters.

[0054] Therefore, in a halftone image with a tone value that is one greater than the first switching threshold, clusters of dots that are gathered within the periodic region are spaced at the same intervals in the row and column directions as the periodic region 83 (see the clusters of determined elements 82 in FIG. 12). A similar characteristic is also present in halftone images with tone values ​​equal to or less than the first switching threshold. In this way, the threshold matrix 8 has the same characteristics as an AM screen in which halftone dots are periodically arranged. As a result, areas in the original image 70 that are made up of pixels with tone values ​​equal to or less than the tone value that is one greater than the first switching threshold are designated as highlight areas, and in areas of the printed image that correspond to these highlight areas (hereinafter also referred to as "highlight areas"), the visual graininess is improved compared to an FM screen.

[0055] Furthermore, within the periodic regions of the halftone dot image, dots are prevented from being adjacent to each other in all four directions (neighborhoods), and dots are adjacent diagonally, resulting in a large distance between dots. As a result, beading, in which ink forming adjacent dots aggregates into beads, is suppressed in highlight regions of the printed image. Furthermore, on the printing paper 9, dots spread due to ink bleeding, but because the distance between adjacent dots is large, overlapping areas in highlight regions are reduced. As a result, it is possible to efficiently change density in gradation changes on the highlight side (i.e., changes in gradation values ​​below the above gradation value), which is known as efficient dot gain.

[0056] In generating halftone dot images, when clusters of uniformly shaped dots are periodically arranged, as in an AM screen, variations in the ink ejection direction or amount from the multiple ejection ports in the head of the ejection unit 2 (also known as printing errors, hereinafter referred to as "variations in dot position, etc.") can result in noticeable streaks of density unevenness in the printed image. In contrast, in the threshold matrix 8, the target elements 81a to which thresholds are assigned in the highlight-side threshold assignment process are identified using randomness similar to that of an FM screen. The arrangement of the determined elements 82 (the outer shapes of the clusters of determined elements 82) in each periodic region 83 is not uniform (see FIG. 11). This makes it possible to suppress density unevenness due to variations in dot position, etc. Furthermore, the occurrence of moire can also be suppressed in printed images of multiple color components.

[0057] Here, the first switching threshold will be explained. Because the number of target elements 81a in the periodic region 83 in the matrix space 80 is approximately 25% of the total number of elements, the first switching threshold set in the highlight-side threshold allocation process is set to a value equal to or less than 25% of the entire threshold range. To improve graininess in the highlight region, it is preferable that dot clusters of a certain size are formed in each periodic region of the halftone dot image. From this perspective, the first switching threshold is preferably set to a value equal to or greater than 15% of the entire threshold range, and more preferably equal to or greater than 18% of the entire threshold range.

[0058] On the other hand, when a first switching threshold is set in the highlight-side threshold allocation process such that all target elements 81a in each periodic region 83 become determined elements 82, clusters of dots with the same outline are formed in all periodic regions in a halftone image with a tone value that is one greater than the first switching threshold. In this case, variations in dot position, etc., can make density unevenness, etc., more noticeable in the printed image. Furthermore, moiré may occur in printed images of multiple color components. To suppress density unevenness, moiré, etc., in the printed image, it is preferable to terminate the highlight-side threshold allocation process before all target elements 81a in each periodic region 83 become determined elements 82. From this perspective, the first switching threshold is preferably a value less than 25% of the entire threshold range, and more preferably a value equal to or less than 23% of the entire threshold range. As will be described later, after the highlight-side threshold allocation process, elements 81 in regions other than the periodic region 83 are preferentially set as determined elements 82, and fluctuations are imparted to the arrangement of the determined elements 82. Depending on the application of the printing device 1, the first threshold may be set to a value less than 25% of the entire threshold range, and more preferably, to a value less than 23% of the entire threshold range. 1 The switching threshold may be 25% of the total threshold range.

[0059] When the highlight-side threshold allocation process is completed, the intermediate processing unit 52 of the calculation unit 50 performs a process of assigning a threshold value between the highlight-side threshold value and the shadow-side threshold value to the elements 81 of the matrix space 80 (hereinafter referred to as "intermediate threshold allocation process") (FIG. 6: step S23). FIG. 13 is a diagram showing the flow of the intermediate threshold allocation process. First, the intermediate processing unit 52 cancels the setting of the periodic region 83 in the matrix space 80 (step S231).

[0060] Next, thresholds ranging from the next threshold after the first switching threshold (the first switching threshold plus 1) to a predetermined second switching threshold are assigned to target elements 81a among the remaining elements 81 to which no threshold has been assigned. Specifically, taking into consideration the repeated application of the threshold matrix 8 when halftoning the original image 70, the target element 81a for which a threshold has not yet been determined and which has the greatest distance from all of the determined elements 82 is identified from the entire matrix space 80. Then, the threshold after the first switching threshold is set as the current threshold, and the current threshold is assigned to the target element 81a (step S232).

[0061] In this case, since the current threshold has not yet been assigned to the set number of elements 81, the current threshold is determined as the next threshold (step S233). Next, the current threshold is compared with the second switching threshold described above. The second switching threshold is greater than the first switching threshold and is a value that is 50% or less of the entire threshold range (127 or less in this processing example). The second switching threshold is preferably a value that is 35% or more of the entire threshold range, and more preferably a value that is 40% or more of the entire threshold range (the reason for this will be explained later). Since the current threshold is less than or equal to the second switching threshold (step S234), the process returns to step S232, and the current threshold is assigned to the target element 81a that has the greatest distance from all of the determined elements 82.

[0062] The above step S232 is repeated until the current threshold is assigned to the set number of elements 81 (steps S233, S234). When the current threshold is assigned to the set number of elements 81, a value obtained by adding 1 to the current threshold is determined as the next threshold (i.e., the next current threshold) (step S233). Then, the above step S232 is repeated until the current threshold is assigned to the set number of elements 81 (steps S233, S234). As described above, when the current threshold is assigned to the set number of elements 81, the process of assigning the current threshold to the target element 81a having the largest distance to all of the determined elements 82 is repeated while determining a value obtained by adding 1 to the current threshold as the next threshold (steps S232 to S234).

[0063] Figure 14 shows a portion of a matrix space 80 during the intermediate threshold assignment process. To aid understanding, in Figure 14, a periodic region 83 for which the setting has already been cancelled is indicated by a thick dashed line. As described above, in the highlight-side threshold assignment process, determined elements 82 exist only in the periodic region 83. Therefore, in the intermediate threshold assignment process, thresholds are more likely to be assigned to target elements 81a that exist in areas other than the periodic region 83, that is, target elements 81a in areas other than the periodic region 83 are given priority as determined elements 82.

[0064] When the current threshold value becomes larger than the second switching threshold value in the repetition of step S232 (Yes in step S234), the intermediate threshold allocation process ends. At the end of the intermediate threshold allocation process, a large number of determined elements 82 are discretely arranged in areas other than the periodic area 83. Therefore, in a halftone image with a gradation value that is 1 greater than the second switching threshold value, the periodic arrangement of clusters of dots (see the clusters of determined elements 82 in FIG. 12) becomes weaker in a halftone image with a gradation value that is 1 greater than the first switching threshold value.

[0065] Here, the second switching threshold will be explained. Because the number of target elements 81a in the matrix space 80 is 50% of the total number of elements, the second switching threshold set in the intermediate threshold allocation process is set to a value equal to or less than 50% of the entire threshold range. In areas other than the highlight areas of a halftone dot image, it is preferable to form a large number of dots in areas other than the periodic areas in order to reduce the influence of clusters of dots arranged in periodic areas (i.e., to cancel the characteristics of the AM screen) (see FIG. 14). From this perspective, the second switching threshold is preferably a value equal to or greater than 35% of the entire threshold range, and more preferably a value equal to or greater than 40% of the entire threshold range.

[0066] On the other hand, when dots 62 are arranged in a checkerboard pattern in a halftone dot image 60 obtained using the threshold matrix 8, as shown in FIG. 15, variations in dot position, etc., can cause streaks of density unevenness to become noticeable in the printed image. Therefore, to minimize the impact of variations in dot position, etc., on the printed image, it is preferable to terminate the intermediate threshold assignment process before all target elements 81a become determined elements 82. From this perspective, the second switching threshold is preferably a value less than 50% of the entire threshold range, and more preferably a value equal to or less than 48% of the entire threshold range. As will be described later, after the intermediate threshold assignment process, elements 81 other than the target elements 81a are also designated as determined elements 82, providing further variation in the arrangement of the determined elements 82. Depending on the application of the printing device 1, the second switching threshold may be a value equal to 50% of the entire threshold range.

[0067] When the intermediate threshold allocation process is completed, the shadow processing unit 53 of the calculation unit 50 performs a process of allocating a shadow side threshold to an element 81 in the matrix space 80 (hereinafter referred to as the "shadow side threshold allocation process") (FIG. 6: step S24). FIG. 16 is a diagram showing the flow of the shadow side threshold allocation process. First, the shadow processing unit 53 cancels the setting of the target element 81a in the matrix space 80 (step S241).

[0068] Next, thresholds are assigned to the remaining elements 81 to which no threshold has been assigned, in order from the threshold next to the second switching threshold (the second switching threshold plus 1) to the threshold furthest from the shadow side. Specifically, taking into consideration the repeated application of the threshold matrix 8 when halftoning the original image 70, an element 81 for which a threshold has not been determined and which is the greatest distance from all determined elements 82 is identified from the entire matrix space 80. Then, the threshold next to the second switching threshold is set as the current threshold, and the current threshold is assigned to the element 81 (step S242).

[0069] In this case, since the current threshold has not yet been assigned to the set number of elements 81, the current threshold is determined as the next threshold (step S243). Next, the current threshold is compared with the shadow-most threshold described above. The shadow-most threshold is 100% of the entire threshold range and is sufficiently greater than the second switching threshold. Since the current threshold is equal to or less than the shadow-most threshold (step S244), the process returns to step S242, and the current threshold is assigned to the element 81 with the greatest distance from all of the determined elements 82.

[0070] The above step S242 is repeated until the current threshold is assigned to the set number of elements 81 (Yes in step S244). When the current threshold is assigned to the set number of elements 81, the value obtained by adding 1 to the current threshold is determined as the next threshold (i.e., the next current threshold) (step S243). Then, the above step is repeated until the current threshold is assigned to the set number of elements 81. S242 As described above, when the current threshold value has been assigned to the set number of elements 81, the process of assigning the current threshold value to the element 81 with the longest distance from all the determined elements 82 is repeated while determining the value obtained by adding 1 to the current threshold value as the next threshold value (steps S242 to S244).

[0071] 17 and 18 are diagrams showing a portion of a halftone dot image 60 obtained using the threshold matrix 8. Fig. 17 shows a halftone dot image 60 with a gradation value of 128, which is the same as the arrangement of determined elements 82 at a stage in the shadow-side threshold assignment process when the current threshold is 127. Fig. 18 shows a halftone dot image 60 with a gradation value of 217, which is the same as the arrangement of determined elements 82 at a stage in the shadow-side threshold assignment process when the current threshold is 216.

[0072] In generating the threshold matrix 8 used to create the halftone dot image 60 of FIG. 17, the second switching threshold is set to 114. Therefore, when the current threshold is 127, many elements 81 other than the target element 81a are set to determined elements 82, adding the randomness of the FM screen. As a result, in the halftone dot image 60 of FIG. 17, the dots 62 are not arranged in a checkerboard pattern like the halftone dot image 60 of FIG. 15, but rather have a fluctuation in their arrangement. This makes streaky density variations less noticeable in the printed image, even when there is variation in dot position, etc. In practice, the shadow-side threshold allocation process performs the same process as generating an FM screen from the threshold next to the second switching threshold to the threshold closest to the shadow side, ensuring the detailed reproducibility that is a characteristic of FM screens.

[0073] In the repetition of step S242, when the current threshold value becomes larger than the threshold value on the shadow side (steps S243 and S244), the shadow side threshold value assignment process ends. This results in a threshold value matrix 8 in which threshold values ​​are assigned to all elements 81. The threshold value matrix 8 is output to the main body control unit 4 in FIG. 3 and stored in the matrix storage unit 42.

[0074] As described above, in steps S223 to S228 of the highlight-side threshold assignment process, thresholds ranging from the most highlight-side threshold to the first switching threshold are assigned in order to target elements 81a included in multiple periodic regions 83. Therefore, when multiple halftone images ranging from the most highlight-side gradation value to a switching gradation value that is just one greater than the first switching threshold are generated using threshold matrix 8, dots are formed in the multiple halftone images only at every other target pixel in the row and column directions in multiple periodic regions that are periodically and uniformly distributed and each include multiple pixels.

[0075] Furthermore, in the above processing, under the condition that the number of determined elements 82 in the multiple periodic regions 83 is approximately the same and that each determined element 82 in each periodic region 83 is located near one of the determined elements 82 (in the above example, adjacent to one of the determined elements 82), a target element 81a whose distance to all of the determined elements 82 is greatest is identified within the multiple periodic regions 83. Therefore, the position at which a dot is added from a halftone image of one gradation value to a halftone image of the next gradation value in the multiple halftone images becomes the target pixel whose distance to all existing dots is maximum under the condition that the number of dots in the multiple periodic regions is approximately the same in each halftone image and that each dot is located near one of the dots in each periodic region.

[0076] In steps S232 to S234 of the intermediate threshold assignment process, thresholds from the next threshold after the first switching threshold to the second switching threshold are assigned in order to target elements 81a among the remaining elements 81 to which no threshold has been assigned. Therefore, when multiple halftone images ranging from the next gradation value after the switching gradation value to another switching gradation value that is just one greater than the second switching threshold are generated using the threshold matrix 8, dots are formed only at target pixels that exist every other pixel in the row and column directions in the multiple halftone images. Furthermore, in the above process, the target element 81a that is the greatest distance from all determined elements 82 is identified. Therefore, the position where a dot is to be added from a halftone image of one gradation value in the multiple halftone images to a halftone image of the next gradation value is the target pixel that is the greatest distance from all existing dots.

[0077] In steps S242 to S244 of the shadow-side threshold allocation process, thresholds from the threshold next to the second switching threshold to the threshold furthest on the shadow side are assigned to the remaining elements 81 to which no threshold has been assigned. This process also identifies the element 81 that is the greatest distance from all of the determined elements 82. Therefore, when multiple halftone images ranging from the next gradation value after the other switching gradation value to the gradation value furthest on the shadow side are generated using the threshold matrix 8, the position at which a dot is added from the halftone image of one gradation value to the halftone image of the next gradation value is the pixel that is the greatest distance from all of the existing dots.

[0078] As described above, the threshold matrices 8 for the K, C, M, and Y color components are generated by the process of FIG. 6, but it is preferable that the threshold matrices 8 for these color components are different from each other. Here, attention is focused on two of the multiple color components, and these are referred to as the "first color component" and the "second color component." In a first method for making the threshold matrices 8 for multiple color components different from each other, in step S222 of FIG. 7, multiple periodic regions 83 are set in the matrix space 80 so that the arrangement of the periodic regions 83 for the first color component differs from the arrangement of the periodic regions 83 for the second color component.

[0079] 19A and 19B are diagrams illustrating multiple periodic regions 83 set in a matrix space 80. In FIGS. 19A and 19B, the multiple periodic regions 83 when generating the threshold matrix 8 for the first color component are indicated by solid lines, and the multiple periodic regions 83 when generating the threshold matrix 8 for the second color component are indicated by dashed lines. In the example of FIG. 19A, the periodic regions 83 for the first color component and the periodic regions 83 for the second color component are arranged so that they do not overlap. In the example of FIG. 19B, the periodic regions 83 for the first color component and the periodic regions 83 for the second color component are arranged so that they partially overlap. In this way, by shifting the arrangement of the periodic regions 83 for each color component, it is possible to make the threshold matrices 8 for multiple color components different from one another. As a result, it is possible to suppress density unevenness due to variations in dot position, etc., compared to when the threshold matrices 8 for multiple color components are the same.

[0080] In the second method for making the threshold matrices 8 for multiple color components different from one another, when generating the threshold matrix 8 for the second color component, the positions of the target elements 81a to which thresholds are assigned in the first step S225 of Fig. 7 are determined to be different from the positions at which the threshold matrix 8 for the first color component was generated. As a result, the arrangement of the target elements 81a to which thresholds are sequentially assigned in the repetition of steps S223, S224, and S228 (S225) differs between the matrix space 80 for the first color component and the matrix space 80 for the second color component.

[0081] 20 and 21 show portions of a halftone dot image 60 obtained using a threshold matrix 8 for multiple color components. From left to right, the halftone dot images 60 are shown for K, C, M, and Y, respectively. FIG. 20 shows a halftone dot image 60 with gradation values ​​greater than the first switching threshold and less than the second switching threshold, and the arrangement is the same as that of determined elements 82 at a certain stage in the intermediate threshold assignment process. FIG. 21 shows a halftone dot image 60 with gradation values ​​greater than the second switching threshold, and the arrangement is the same as that of determined elements 82 at a certain stage in the shadow-side threshold assignment process. The threshold matrices 8 for multiple color components used to create the halftone dot images 60 in FIGS. 20 and 21 differ from each other due to the adoption of the first and second techniques described above.

[0082] In the halftone dot images (semi-transparent halftone dot images 60) of K, C, M, and Y shown in Figures 20 and 21, the dot arrangements are different in the highlight tone range where dots are formed in periodic regions, the midtone range where dots are preferentially formed in regions outside the periodic regions, and the shadow tone range where processing similar to FM screening is performed. In the halftone dot images of K, C, M, and Y, the high randomness in the dot arrangement makes it possible to suppress density unevenness caused by external disturbances or variations in dot position. The threshold matrices 8 for multiple color components may be made different from each other by changing the weighting coefficients included in the formula used to identify the element 81 with the greatest distance from all determined elements 82, or by changing the formula itself.

[0083] Next, a comparison is made between a printed image printed on printing paper 9 using threshold matrix 8 and a printed image printed on printing paper 9 using a threshold matrix for a general FM screen. FIG. 22 is a photograph showing the highlight and shadow regions of a printed image printed using threshold matrix 8. FIG. 23 is a photograph showing the highlight and shadow regions of a comparative printed image printed using a threshold matrix for a general FM screen. In FIGS. 22 and 23, the left side shows the highlight region, and the right side shows the shadow region. As described above, the highlight region is a region corresponding to a region in the original image 70 that is made up of pixels with gradation values ​​on the highlight side. Furthermore, the shadow region is a region corresponding to a region in the original image 70 that is made up of pixels with gradation values ​​on the shadow side.

[0084] In the highlight region on the left side of Figure 23, many dots are arranged irregularly, which tends to create a visually grainy appearance and worsens graininess. In contrast, in the highlight region on the left side of Figure 22, clusters of dots are arranged periodically, which tends to create a less grainy appearance and improves graininess. The shadow region on the right side of Figure 22 and the shadow region on the right side of Figure 23 are roughly equivalent.

[0085] Fig. 24 is a photograph showing a portion of a printed image printed using threshold matrix 8. Fig. 25 is a photograph showing a portion of a printed image of a comparative example printed using a threshold matrix for a general FM screen. Figs. 24 and 25 show the character area. In both Figs. 24 and 25, there is no line breakage or the like, and the same level of detail reproducibility is achieved.

[0086] As described above, the method for generating the threshold matrix 8 includes the steps of preparing a matrix space 80 (step S21), setting multiple periodic regions 83 in the matrix space 80 (step S222), setting every other element 81 in the row and column directions in each periodic region 83 as a target element 81a (step S221), assigning thresholds ranging from the most highlighted threshold to a predetermined switching threshold (first switching threshold) to the target elements 81a included in the multiple periodic regions 83 in order (steps S223 to S228), and, after steps S223 to S228, assigning the remaining thresholds in order to the remaining elements 81 to which no threshold has been assigned (steps S23, S24). Furthermore, when assigning each threshold value in steps S223 to S228, the element 81 to which a threshold value has been assigned is considered to be a determined element 82, and under the condition that the number of determined elements 82 in the multiple periodic regions 83 is approximately the same and each determined element 82 in each periodic region 83 is located in the vicinity of another determined element 82, the target element 81a that is the longest distance from all determined elements 82 is identified within the multiple periodic regions 83 and assigned the threshold value.

[0087] As a result, in the highlight regions of the printed image, dot clusters are arranged periodically, similar to AM screens, making it less likely for the image to appear rough and improving graininess. Also, in the highlight regions of the printed image, dots (of the same color) are prevented from being adjacent to each other vertically and horizontally, suppressing the occurrence of beading. Furthermore, areas where dots overlap in the highlight regions are reduced, allowing for efficient density changes in gradation changes on the highlight side. In multiple periodic regions of the printed image, the contours of dot clusters tend to differ, suppressing the occurrence of density unevenness due to variations in dot position, etc.

[0088] Preferably, in steps S223 to S228, target elements 81a adjacent to any of the determined elements 82 are identified and assigned a threshold value. This prevents the distance between dots forming a cluster of dots from becoming too large, preventing the cluster from being recognized as a single cluster of dots, and more reliably improves graininess in highlight areas of the printed image.

[0089] Preferably, when assigning each threshold value in steps S23 and S24, the element 81 with the greatest distance to all determined elements 82 is identified and assigned the threshold value. This makes it possible to improve the graininess in highlight areas of a printed image, similar to the AM screen, while also improving the detail reproduction that is reduced in the AM screen.

[0090] Preferably, in step S221, every other element 81 in the row and column directions in the matrix space 80 is set as a target element 81a. Furthermore, steps S23 and S24 above include the steps of assigning thresholds to the target elements 81a among the remaining elements 81, in order, from the threshold next to the switching threshold to another switching threshold (second switching threshold) (steps S232 to S234), and, after steps S232 to S234, assigning thresholds to the remaining elements 81 to which no threshold has been assigned, in order, from the threshold next to the other switching threshold to the threshold furthest from the shadow side (steps S242 to S244). This prevents dots from being adjacent to each other vertically and horizontally over a wide range of gradations, making it possible to suppress beading and improve dot gain efficiency.

[0091] Preferably, the threshold matrix 8 for the first color component and the threshold matrix 8 for the second color component are generated by repeating the above-described method for generating the threshold matrix 8. Furthermore, in steps S223 to S228 when generating the threshold matrix 8 for the first color component, the position of the target element 81a to which the most highlight-side threshold is initially assigned differs from that when generating the threshold matrix 8 for the second color component. This makes it possible to make the threshold matrix 8 for the first color component different from the threshold matrix 8 for the second color component, thereby making it possible to prevent density unevenness and the like from occurring in the printed image due to variations in dot positions, etc.

[0092] Furthermore, the arrangement of the periodic regions 83 when generating the threshold matrix 8 for the first color component may be different from that when generating the threshold matrix 8 for the second color component. In this case as well, the threshold matrix 8 for the first color component and the threshold matrix 8 for the second color component can be made different, making it possible to prevent density unevenness and the like from occurring due to variations in dot positions, etc.

[0093] The image data generation method includes a step (step S11) of preparing a threshold matrix 8 generated by the above generation method, and a step (step S12) of generating halftone image data by halftoning the original image 70 by comparing the original image 70 with the threshold matrix 8. This makes it possible to generate a halftone image with improved graininess in highlight areas.

[0094] In the threshold matrix 8, when multiple halftone dot images are generated, ranging from the most highlight-side gradation value to a predetermined switching gradation value, dots are formed only at target pixels that exist every other pixel in the row and column directions in multiple periodic regions that are periodically and uniformly distributed in the multiple halftone dot images and each contain multiple pixels. Furthermore, the position at which a dot is added from a halftone dot image of one gradation value to a halftone dot image of the next gradation value in the multiple halftone dot images is the target pixel that is the longest distance from all existing dots, under the conditions that the number of dots in the multiple periodic regions in each halftone dot image is approximately the same and each dot is located near another dot in each periodic region. This improves graininess in highlight regions in the printed image and efficiently changes density when changing gradations on the highlight side.

[0095] The image data generating device (main body control unit 4 in the above example) includes a matrix storage unit 42 that stores the threshold matrix 8, and an image data generating unit (comparator 43 in the above example) that generates halftone image data by halftone-doting the original image 70 by comparing the multi-tone original image 70 with the threshold matrix 8. This makes it possible to generate a halftone image with improved graininess in highlight areas.

[0096] 26A to 26F are diagrams showing other examples of the periodic region 83. In FIGS. 26A to 26F, the periodic region 83 is indicated by parallel diagonal lines. In the example described above, the periodic region 83 is a square region consisting of 5 rows and 5 columns of elements 81 (see FIG. 26A), but as shown in FIG. 26B, the periodic region 83 may be a smaller region. For example, when using ink that easily bleeds or a substrate that easily bleeds ink, it is preferable to employ a large periodic region 83 in order to prevent dots from joining together between adjacent periodic regions in the printed image.

[0097] Furthermore, when the resolution (dpi) in the width direction and the movement direction of the printing device 1 differs, the number of elements 81 in the row direction and the number of elements 81 in the column direction in each periodic region 83 may differ, as shown in FIGS. 26C and 26D. In the example of FIG. 26C, the resolution in the movement direction corresponding to the column direction is approximately half the resolution in the width direction corresponding to the row direction, so the number of elements 81 in the column direction in each periodic region 83 is approximately half the number of elements 81 in the row direction. In the example of FIG. 26D, the resolution in the width direction is approximately half the resolution in the movement direction, so the number of elements 81 in the row direction in each periodic region 83 is approximately half the number of elements 81 in the column direction. In this way, by adjusting the aspect ratio of the periodic region 83 according to the resolution in the width direction and the movement direction, it is possible to prevent the outline of dot clusters in highlight regions in the printed image from becoming flat. Note that the number of elements 81 in each direction in the periodic region 83 is not particularly limited, but is, for example, 3 to 8.

[0098] As long as the periodic regions 83 are periodically arranged and uniformly distributed, gaps (elements 81) may be provided between diagonally adjacent periodic regions 83, as shown in FIG. 26E. Alternatively, as shown in FIG. 26F, a plurality of periodic regions 83 aligned in a direction oblique to the row direction may be defined as a periodic region group, and the plurality of periodic region groups may be arranged in the column direction. In this way, the plurality of periodic regions 83 are arranged at regular intervals in each of two directions intersecting each other. The shape of the periodic regions 83 is not limited to a rectangle, and may be any shape, such as a rhombus.

[0099] FIG. 27 is a diagram showing a matrix space 80 in another processing example. In this processing example, target elements 81a are set only in periodic regions 83 (FIG. 7: steps S221 and S222). At this time, in two periodic regions 83 adjacent to each other in the diagonal direction, the arrangement of the target elements 81a is shifted so that the target elements 81a do not contiguous with each other. That is, in the two periodic regions 83, at least one of the two elements 81 adjacent to each other in the diagonal direction is a non-target element. This more reliably prevents clusters of dots from connecting with each other in highlight regions of the print image. The processing in steps S223 to S228, in which thresholds are assigned to target elements 81a included in multiple periodic regions 83, is the same as in the processing example described above.

[0100] As described above, in generating the threshold matrix 8, it is sufficient to set as target elements 81a every other element 81 in the row and column directions at least in each periodic region 83. This allows for efficient change in density during tone changes on the highlight side.

[0101] In this processing example, where no target elements 81a are set in areas other than the periodic area 83, after the highlight-side threshold assignment process is completed, the remaining thresholds are assigned in order to the remaining elements 81 to which no thresholds have been assigned. Specifically, as in the shadow-side threshold assignment process described above, when assigning each threshold, the element 81 that is the greatest distance from all of the determined elements 82 is identified and assigned that threshold, taking into consideration the repeated application of the threshold matrix 8 used when halftoning the original image 70.

[0102] The generation of the threshold matrix 8 and the generation of image data using the threshold matrix 8 can be modified in various ways.

[0103] In the above embodiment, when assigning each threshold value in steps S223 to S228, a condition is set that each determined element 82 is adjacent to another determined element 82 in each periodic region 83, but each determined element 82 may be located in the vicinity of another determined element 82. In the example of FIG. 28, target elements 81a (target elements 81a surrounded by dashed lines in FIG. 28) adjacent to the eight neighboring elements 81 of the central determined element 82 are also included in the candidates for target elements 81a to which the current threshold value should be assigned in step S228. In other words, when a threshold value is assigned, the target elements 81a adjacent to the eight neighboring elements 81 of the determined element 82 also become determined elements 82 located in the vicinity of the determined element 82. It can be said that the dots of the print image corresponding to these determined elements 82 also form a single dot cluster.

[0104] 7, the setting of the target element 81a in step S221 and the setting of the multiple periodic regions 83 in step S222 may be performed in any order. Furthermore, the setting of the target element 81a and the periodic regions 83 does not necessarily have to be performed explicitly, and the target element 81a and the periodic regions 83 may simply be taken into consideration in the calculation for identifying the element 81 to which a threshold should be assigned. In this case as well, it can be said that the setting of the target element 81a and the setting of the multiple periodic regions 83 have essentially been performed.

[0105] After the highlight side threshold allocation process, in the process of assigning the remaining thresholds to the remaining elements 81 to which thresholds have not been assigned, it is not necessary to identify the element 81 that has the greatest distance from all determined elements 82; the element 81 to which a threshold is to be assigned may be identified using other methods.

[0106] When generating the threshold matrix 8 for multiple color components, the shape or size of the multiple periodic regions 83 when generating the threshold matrix 8 for a first color component may be different from that when generating the threshold matrix 8 for a second color component. In this case, too, by making the threshold matrix 8 for the first color component different from the threshold matrix 8 for the second color component, it is possible to prevent density unevenness and the like from occurring in the printed image.

[0107] In the threshold matrices for a plurality of color components (for example, K, C, M, Y), the threshold matrices for some color components may be other types of threshold matrices, such as a threshold matrix for an FM screen.

[0108] The threshold matrix 8 may be used in other devices related to the formation of printed images, such as electrophotographic printing devices and CTP (Computer To Plate) plate-making devices used in offset printing, etc. By using the threshold matrix 8 in such other devices, it is possible to reduce overlapping of dots in highlight areas of a printed image, improve graininess, and efficiently change density when the gradation changes on the highlight side.

[0109] The configurations in the above-described embodiment and each modification may be combined as appropriate as long as they are not mutually contradictory. [Explanation of symbols]

[0110] 4 Main unit control section 5. Computer 8 Threshold Matrix 42 Matrix storage unit 43 Comparator 60 flat screen images 62 dots 70 original images 80 Matrix Space 81 elements 81a Target element 82 Determined Elements 83 Periodic region 900 Programs S11~S15, S21~S24, S221~S228, S231~S234, S241~S244 steps

Claims

1. 1. A threshold matrix generation method for generating a threshold matrix to be compared with an original image when halftoning the original image with multiple tones, comprising: a) preparing a matrix space which is a set of elements arranged in rows and columns; b) defining a plurality of periodic regions in the matrix space, the periodic regions being periodically arranged and uniformly distributed, each of the periodic regions including a plurality of elements; c) setting every other element in the row direction and the column direction as a target element in at least each periodic region; d) assigning thresholds to target elements included in the plurality of periodic regions in order from the most highlight-side threshold to a predetermined switching threshold; e) after step d), assigning remaining thresholds in order to the remaining elements to which no thresholds have been assigned; Equipped with a threshold matrix generating method characterized in that, when assigning each threshold in step d), an element to which a threshold is assigned is considered to be a determined element, a target element within the plurality of periodic regions that has the greatest distance from all determined elements is identified and assigned each threshold, while taking into consideration repeated application of the threshold matrix when halftoning the original image, under the conditions that the difference between the maximum and minimum number of determined elements in the plurality of periodic regions is 1 or 0, and that each determined element in each periodic region is adjacent to any determined element or adjacent to an element within eight neighborhoods of any determined element.

2. 2. The threshold matrix generating method according to claim 1, In the step d), a target element adjacent to any determined element is identified and a threshold value is assigned to the target element.

3. 3. The threshold matrix generation method according to claim 1, further comprising: a threshold matrix generating method characterized in that, when assigning each threshold in the step e), an element having the maximum distance from all determined elements is identified and assigned each threshold, while taking into consideration the repeated application of the threshold matrix when halftoning the original image.

4. 4. The threshold matrix generating method according to claim 3, In the step c), elements existing every other in the row direction and the column direction in the matrix space are set as target elements; The step e) e1) assigning thresholds to target elements among the remaining elements in order from the threshold next to the switching threshold to another switching threshold; e2) after step e1), assigning thresholds to the remaining elements to which no threshold has been assigned, in order from the threshold next to the other switching threshold to the threshold closest to the shadow side; A threshold matrix generating method comprising:

5. 5. The threshold matrix generating method according to claim 1, further comprising: By repeating the steps a) to e), a threshold matrix for the first color component and a threshold matrix for the second color component are generated; a threshold matrix generating method, characterized in that the position of the target element to which the most highlight-side threshold value is first assigned in step d) when generating the threshold matrix for the first color component is different from that when generating the threshold matrix for the second color component.

6. 6. The threshold matrix generating method according to claim 1, further comprising: By repeating the steps a) to e), a threshold matrix for the first color component and a threshold matrix for the second color component are generated; A threshold matrix generating method, wherein the arrangement, shape or size of the plurality of periodic regions when generating the threshold matrix for the first color component is different from that when generating the threshold matrix for the second color component.

7. An image data generation method for generating image data, comprising: a step of preparing a threshold matrix generated by the threshold matrix generating method according to any one of claims 1 to 6; a step of comparing a multi-tone original image with the threshold matrix to generate halftone image data by halftoning the original image; An image data generating method comprising:

8. A program for causing a computer to generate a threshold matrix to be compared with an original image having multiple tones when the original image is halftone-coded, the program being executed by the computer by the computer comprising: a) preparing a matrix space which is a set of elements arranged in rows and columns; b) defining a plurality of periodic regions in the matrix space, the periodic regions being periodically arranged and uniformly distributed, each of the periodic regions including a plurality of elements; c) setting every other element in the row direction and the column direction as a target element in at least each periodic region; d) assigning thresholds to target elements included in the plurality of periodic regions in order from the most highlight-side threshold to a predetermined switching threshold; e) after step d), assigning remaining thresholds in order to the remaining elements to which no thresholds have been assigned; Execute a program for assigning each threshold in step d) by regarding elements to which thresholds have been assigned as determined elements, wherein, under the conditions that the difference between the maximum and minimum number of determined elements in the plurality of periodic regions is 1 or 0, and that each determined element in each periodic region is adjacent to any determined element or adjacent to an element within eight neighborhoods of any determined element, the program identifies within the plurality of periodic regions a target element that has the greatest distance from all determined elements, while taking into account the repeated application of the threshold matrix when halftoning the original image, and assigns each threshold to the target element.

9. A threshold matrix to be compared with an original image when halftoning the original image having multiple tones, When a plurality of halftone dot images ranging from the most highlight-side gradation value to a predetermined switching gradation value are generated, dots are formed only at target pixels that exist every other pixel in the row and column directions in a plurality of periodic regions that are periodically arranged and uniformly distributed in the plurality of halftone dot images, each of which includes a plurality of pixels; A threshold matrix characterized in that the position at which a dot is added from a halftone image of one gradation value to a halftone image of the next gradation value in the plurality of halftone images is a target pixel where the difference between the maximum and minimum numbers of dots in the plurality of periodic regions in each halftone image is 1 or 0, and where each dot in each periodic region is adjacent to any dot or adjacent to the eight nearest pixels of any dot, and where the distance to all existing dots is the greatest.

10. An image data generating device that generates image data, a matrix storage unit that stores the threshold matrix according to claim 9; an image data generating unit that generates halftone image data by comparing a multi-tone original image with the threshold matrix, and halftone-dots the original image; An image data generating device comprising:

Images