JP5891772B2 - Printing apparatus and printed matter production method - Google Patents

Description

  The present invention relates to a printing technique for printing an image using dots.

  Conventionally, a technique for reproducing a multi-tone image by recording one kind or several kinds of dots on a printing medium is used in a printing apparatus such as a printer. In recent years, such multi-gradation technology has been remarkably developed, and there are about two types of inks of several hues such as cyan (C), magenta (M), yellow (Y), and black (K). By combining dots and controlling their distribution, it is possible to form a so-called photographic image. Such dot arrangement includes a dot concentration type represented by a halftone dot type that concentrates dots and a dot dispersion type in which dots are distributed as much as possible. In the latter case, with the development of technology that analyzes the distribution of dots in the spatial frequency region, the dot distribution now has a noise characteristic that minimizes components below a predetermined frequency in the spatial frequency region. Therefore, it is known that the image quality can be improved.

  A typical such noise characteristic is a blue noise characteristic. Blue noise means, for example, a characteristic that the spatial frequency of an image in which dots are uniformly formed in order to reproduce an image having a constant gradation value has almost no component below a predetermined frequency. The human eye is sensitive to low frequency components below a certain level, but the visibility of high frequency components is not high. For this reason, an image having such a blue noise characteristic is felt to be smooth and high quality. The following patent technology document 1 etc. are known as what disclosed the formation technique of the image which has a blue noise characteristic.

US Pat. No. 5,341,228

  One method for forming dots on a print medium is a serial printer that forms dots while reciprocating a print head in the width direction of the print medium (hereinafter, head movement in this direction is referred to as main scanning). . Among such serial printers, in an inkjet type printer, dots are formed by ejecting ink droplets from a nozzle onto a print medium. For discharging ink droplets, there are known a method using deformation generated when a voltage is applied to a piezo element, a method using bubbles generated by heating ink, and the like. In either method, the formation of an image using dots requires a plurality of print head main scans. Therefore, with the aim of improving the printing speed, bidirectional printing (also referred to as Bi-D printing) in which an image is formed by combining dots formed by the forward movement of the print head and dots formed by the backward movement is formed. With the aim of improving image quality, techniques such as multi-pass printing have been developed in which each raster is completed by a plurality of main scans of the print head.

  In such a technique for forming an image by a plurality of main scans of the print head, image quality deteriorates when the positions of dots formed by one main scan and the positions of dots formed by another main scan are shifted. It is known that it will. This problem appears as a conspicuous conversely in a printing apparatus that determines the optimum dot formation position, assuming that there is no deviation in the dot formation position. FIG. 28 is an explanatory diagram showing an example of dot arrangement when a deviation occurs in the forward and backward dot formation positions in a printer that performs printing by dot-concentrated dot formation. In FIG. 28, “◯” indicates the formation position of dots formed during forward movement, and “●” indicates the formation position of dots formed during backward movement. FIG. 28A shows a case where the dot formation position is not displaced during forward movement and backward movement, and the dot formation position is completely adjusted (hereinafter referred to as “positional relationship of zero displacement”). Yes. Note that the dot size in FIG. 28 is the smallest perfect circle that completely covers each pixel assumed to be a rectangle according to the vertical and horizontal resolution.

  FIG. 28B shows how dots are formed when the dot formation position at the time of forward movement and at the time of backward movement is shifted by one pixel in the main scanning direction due to the positional relationship of zero shift. Similarly, FIG. 28C shows a case where the pixel is shifted by 2 pixels in the main scanning direction, FIG. 28D is a pixel where the pixel is shifted by 3 pixels in the main scanning direction, and FIG. Each case is shown shifted. As can be seen from each figure, as the dot formation position in the main scanning direction deviates from the positional relationship of zero deviation between forward movement and backward movement, the dot concentration type dot arrangement collapses and each dot is independent. Thus, it can be seen that the ratio of covering one pixel increases, and even when the same number of dots are arranged, the ratio of the dots covering the surface of the print medium (hereinafter referred to as coverage) increases. When the dot coverage changes, the brightness (brightness, density, reflectance, etc.) and hue (hue and saturation) of the image change. FIG. 28 exemplifies a case where dots are alternately formed for each raster during forward movement and backward movement. However, dots are alternately formed during forward movement and backward movement. Even when dots are formed on a checkered pattern (checkered pattern), a shift in the dot formation position due to the reciprocating motion does not change the brightness and color of the image.

  On the other hand, in a printer that uses dot dispersion type dot arrangement, for example, a printer that performs printing with the above-described blue noise characteristics in the dot arrangement, when forming an image with a low gradation value, arrange the dots as close as possible. Is done. However, as described above, when the dot formation positions in the forward movement and the backward movement deviate from the positional relationship of deviation 0, the dot arrangement deviates from the ideal arrangement. In general, in the dot dispersion type dot arrangement, if the dot formation position is shifted, the overlap between dots increases, so that the coverage rate decreases, contrary to the dot concentration type. Even in this case, the brightness and color of the image may change as a result due to the variation in the coverage.

  As described above, in the technique of forming an image by combining dots formed in different main scans, a shift in the dot formation position in each main scan causes a decrease in image quality of the formed image, particularly a gradation value and the like. It will appear as a change in color. The present invention aims to solve such a problem. The displacement of the dot formation position is typically an error in bidirectional printing in which dots are formed respectively when the print head moves forward and backward, but is not necessarily limited thereto. For example, an error caused by a multi-pass printing method in which one raster is formed by a plurality of main scans is also known. This error can occur even if main scanning is limited to unidirectional printing.

  In addition, there is also known a phenomenon in which the landing position of the ink droplet, that is, the dot formation position is shifted due to the bending of the printing medium, for example, the printing paper absorbs the ink and bends (so-called cockling). Naturally, such misalignment of dot formation positions is not limited to printing devices that use ink droplets, but other types of printing, such as thermal transfer printing devices, thermal sublimation types, or so-called line printers with print heads arranged in the width direction of the paper. Even in the apparatus, when dots are formed in the same region and divided into a plurality of pixel groups, this problem may occur.

  SUMMARY An advantage of some aspects of the invention is to solve at least a part of the problems described above, and the invention can be implemented as the following forms or application examples.

Application Example 1: A printing apparatus that performs printing by forming dots on a print medium,
A print head for performing color printing with multi-color dots at a resolution N (N is a natural number of 180 or more) dpi (dot per inch);
When forming dots on the print medium by the print head according to dot data representing the presence or absence of dot formation for each pixel, the dots are formed by dividing the dots into at least first and second pixel groups, A printing unit that prints an image by performing at least part of dot formation by the first and second pixel groups in a common area;
The printing unit performs printing according to the first setting when the speed of the print head with respect to the print medium is lower than a reference speed, and performs printing according to the second setting when the speed of the print head is equal to or higher than the reference speed. Done
In the printing with the second setting, when printing is performed in each of a plurality of states in which the relative positions of the first pixel group and the second pixel group are different, images printed in the plurality of states are printed. The color difference is suppressed to a predetermined value or less.

  In this application example, when the speed of the print head is equal to or higher than the reference speed, that is, when the deviation of the dot formation position is likely to occur, a plurality of different relative positions of the first pixel group and the second pixel group. By suppressing the color difference when printing is performed in each of the states to a predetermined value or less, it is possible to suppress deterioration in image quality.

Application Example 2: The printing apparatus according to Application Example 1,
The second setting is:
(A) When an image to be printed is yellow (RGB value, R = G = 255, B = 0-64, where RGB value is converted into 8 bits), the image is printed in each of the different states. The color difference in the CIE L * a * b * color space of the image is 2 or less as a change in b *,
(B) When the image to be printed is magenta (RGB value, R = B = 255, G = 0-64, where RGB value is converted into 8 bits), printing is performed in each of the different states. The color difference in the CIE L * a * b * color space of the printed image is 0.5 or less as a change in L * or 0.5 or less as a change in a *, and (C) the printed image is cyan (RGB values, G = B = 255, R = 0-64, where RGB values are converted into 8 bits) CIEL * a * b * of the image printed in each of the different states The color difference in the color space satisfies at least one of values of 0.5 or less as a change in L *.

  In this printing apparatus, since the color difference is set in the above range in the L * a * b * color space, the brightness and color tone of the image when the dot formation position is shifted between the plurality of pixel groups. Fluctuations are suppressed.

Application Example 3: The printing apparatus according to Application Example 2,
The plurality of states include a formation position of one dot belonging to the first pixel group and a formation position of a dot belonging to the second pixel group and corresponding to the one dot and adjacent to the one dot. The difference in the distance between dots is 2/720 inch to 5/720 inch.
Even in such a case, fluctuations in the brightness and color of the image are suppressed within the above range.

Application Example 4: The printing apparatus according to any one of Application Example 1 to Application Example 3,
The print head includes dot forming elements arranged at a pitch corresponding to the print resolution in the width direction of the print medium,
The printing unit forms an image using the print head while transporting the print medium in a direction intersecting the width direction of the print head,
The first pixel group includes dots formed by dot forming elements arranged at predetermined positions in the transport direction of the print medium of the print head, and the second pixel group includes the dot of the print head. It includes dots formed by dot forming elements arranged at positions different from the predetermined position in the conveyance direction of the print medium.

  In the printing apparatus according to this application example, the second pixel group is a position different from a predetermined position where a dot forming element that forms dots belonging to the first pixel group is arranged in the transport direction of the print medium of the print head. The dot forming elements arranged in the above are included, and a so-called line printer configuration is provided. Even in such a line printer, the dot formation position may be shifted, but even in such a case, fluctuations in brightness and hue are sufficiently suppressed. In this application example, the speed of the print head with respect to the print medium corresponds to the conveyance speed of the print medium.

Application Example 5: The printing apparatus according to any one of Application Example 1 to Application Example 3,
The print head is relatively movable in a main scanning direction that is a width direction of the print medium, and a plurality of nozzles that eject ink to the print medium are arranged in a direction intersecting the main scan direction. Comprising a plurality of nozzle rows spaced apart in the main scanning direction,
The printing unit repeatedly forms the image by repeating the main scanning and sub-scanning, which is a relative movement of the print medium in a direction crossing the scanning direction,
The first and second pixel groups have any one of the following relationships (a), (b), and (c).
(A) The first pixel group includes dots formed when the print head moves forward in the main scanning direction, and the second pixel group includes dots formed when the print head moves backward in the main scanning direction. including.
(B) A group of pixels in which dots are formed by different nozzle rows among the plurality of nozzle rows.
(C) The first pixel group includes dots formed in one main scan of the print head, and the second pixel group is in a main scan different from the one main scan of the print head. Includes dots to be formed.

  Such a configuration corresponds to various aspects of so-called serial printers, and in such serial printers, it is caused by misalignment of dot formation positions for bidirectional printing and misalignment between nozzle arrays that form dots by ejecting ink droplets. Fluctuations in brightness and hue can be sufficiently suppressed even if there is a shift in the dot formation position to be performed, or a shift in the dot formation position between main scans when a single raster is completed in a plurality of main scans. . In this application example, the speed of the print head relative to the print medium corresponds to the moving speed of the print head by main scanning.

Application Example 6: The printing apparatus according to Application Example 5,
In the first setting, the speed of the print head by the main scanning is 200 cps or more.

In the case of 200 cps (Character Per Second) or more, that is, when the moving speed of the print head is high, the dot formation position is liable to shift, and the present invention is particularly effective. Since the first setting is 200 cps or more, the reference speed and the print head speed in the second setting are also 200 cps or more. Note that 1 cps = 0.1 inch / s (2.54 × 10 ⁻³ m / s). Therefore, 200 cps = 20 inches / s.

Application Example 7: Printed matter production method for producing printed matter by forming dots on a print medium using a print head,
The dot formation is performed at least in the first and second pixel groups, and at least a part of the dot formation by the first and second pixel groups is performed in the common region, so that the resolution N (N is 180 or more natural numbers) dpi (dot per inch), and color printing with multi-colored dots,
When forming dots on the print medium by the print head according to dot data representing the presence or absence of dot formation for each pixel, the dots are formed by dividing the dots into at least first and second pixel groups, An image is printed by performing at least a part of dot formation by the first and second pixel groups in the common area,
When the speed of the print head with respect to the print medium is less than a reference speed, printing is performed according to the first setting. When the speed of the print head is equal to or higher than the reference speed, printing is performed according to the second setting.
In the printing with the second setting, when printing is performed in each of a plurality of states in which the relative positions of the first pixel group and the second pixel group are different, images printed in the plurality of states are printed. The color difference is suppressed to a predetermined value or less.
According to this application example, the same effect as in Application Example 1 can be obtained.

1 is a schematic configuration diagram of a printer 20 according to an embodiment of the present invention. It is explanatory drawing which illustrates the nozzle row of the print head 90 in an Example. It is explanatory drawing which shows the variation of the combination of the dot formed at the time of forward movement, and the dot formed at the time of backward movement. 6 is a flowchart illustrating a printing process in the embodiment. It is explanatory drawing which shows the dot formed at the time of a forward movement and a backward movement, and its combination. It is explanatory drawing which illustrates the case where the shift | offset | difference arises in the dot formation position at the time of a forward movement and a backward movement. It is explanatory drawing which shows the relationship between the deviation | shift amount of a dot formation position, and the variation | change_quantity (DELTA) L of L * about a cyan image. FIG. 5 is an explanatory diagram showing a relationship between a dot formation position shift amount and an L * change amount ΔL and a b * change amount Δb for magenta and yellow images, respectively. It is explanatory drawing which shows the relationship between the deviation | shift amount of a dot formation position, and the variation | change_quantity (DELTA) a of a * about the image of cyan and magenta. It is explanatory drawing which shows the adjacent pixels NR and ND with respect to the pixel of interest OJ. It is a graph which shows the relationship between the dot generation rate k and the pair dot generation rate K. It is a graph which illustrates the coverage rate fluctuation | variation when the shift | offset | difference arises in the dot formation position at the time of a forward movement and a backward movement. Is a graph showing the relationship between deviation from k ² shift amount paired dot generation rate of the dot formation positions in units of pixels. It is a flowchart which shows the production | generation method of a pair pixel control mask. It is explanatory drawing which shows the relationship between the gradation value S and a pair dot target value. It is explanatory drawing which shows an example of the sensitivity characteristic VTF (Visual Transfer Function). 6 is a process diagram showing a flow of a method for generating a first dither mask 61 and a second dither mask 62 used for halftone processing of the printer 20. FIG. FIG. 11 is a process diagram showing the flow of dither mask evaluation processing in the method of generating the first dither mask 61 and the second dither mask 62. It is explanatory drawing which shows the state etc. in which the threshold value was stored in a part of storage element of the 1st dither mask 61 and the 2nd dither mask 62. It is explanatory drawing of a blue noise characteristic and a green noise characteristic. It is explanatory drawing which shows the threshold value arrangement | positioning of the 1st dither mask 61. FIG. It is explanatory drawing which shows the 1st example of the threshold value arrangement | positioning of the 2nd dither mask. It is explanatory drawing which shows the 2nd example of the threshold value arrangement | positioning of the 2nd dither mask. It is explanatory drawing which shows the 3rd example of the threshold value arrangement | positioning of the 2nd dither mask. It is explanatory drawing which shows the 4th example of threshold value arrangement | positioning of the 2nd dither mask. 6 is an explanatory diagram illustrating an example of an effect of a printer. FIG. It is explanatory drawing which shows the threshold value arrangement | positioning of the 2nd dither mask 62 as a modification. It is explanatory drawing which shows that a coverage changes with the shift | offset | difference of a dot formation position in a prior art.

A. First embodiment:
A first embodiment of the present invention will be described.
A-1. Device configuration:
FIG. 1 is a schematic configuration diagram of a printer 20 as a first embodiment of the present invention. The printer 20 is a serial inkjet printer that performs bidirectional printing. As illustrated, the printer 20 includes a mechanism that transports the printing paper P by a paper feed motor 74 and a carriage 80 that moves the carriage 80 by the carriage motor 70. A mechanism for reciprocating in the direction, a mechanism for driving the print head 90 mounted on the carriage 80 to discharge ink and forming dots, and the paper feed motor 74, carriage motor 70, print head 90, and operation panel 99. The control unit 30 is responsible for exchanging signals with the control unit 30. The printer 20 performs printing in either the high image quality mode or the high speed mode as bidirectional printing. The high image quality mode is a mode in which the moving speed of the carriage 80 is set to 200 cps. The high speed mode is a mode in which the moving speed of the carriage 80 is set to 400 cps. Which one is adopted is determined in accordance with an input from the operation panel 99.

  The mechanism for reciprocating the carriage 80 in the axial direction of the platen 75 is installed in parallel with the axis of the platen 75 and is driven endlessly between the slide shaft 73 that holds the carriage 80 slidably and the carriage motor 70. A pulley 72 or the like that stretches the belt 71 is used.

  The carriage 80 contains color inks containing cyan ink (C), magenta ink (M), yellow ink (Y), black ink (K), light cyan ink (Lc), and light magenta ink (Lm), respectively. Ink cartridges 82 to 87 for ink are mounted. In the print head 90 below the carriage 80, nozzle rows corresponding to the above-described color inks are formed. When these ink cartridges 82 to 87 are mounted on the carriage 80 from above, ink can be supplied from each cartridge to the print head 90.

  As shown in FIG. 2, the print head 90 is provided with a nozzle row in which a plurality of nozzles that eject ink droplets are arranged in the sub-scanning direction corresponding to each ink color. The nozzle array pitch R in the nozzle row is an integral multiple of the dot formation pitch (raster interval r), and during printing, the paper is relatively moved in the sub-scanning direction with respect to the print head 90 for each main scanning. By repeating main scanning a plurality of times while moving, each raster is completed, so-called interlaced printing is performed. In addition, so-called overlap printing, in which one raster is completed by a plurality of main scans, can also be performed. For this reason, by combining interlacing and overlapping, each raster or each column is unified with dots formed either when the print head 90 moves forward or backward, so-called alternating dot arrangement of columns ( Printing can be performed with FIG. 3 (A)) or with raster alternate dot arrangement (FIG. 3 (B)). Alternatively, it is also possible to print with a so-called task dot arrangement (FIG. 3C) in which dots formed during forward movement and dots formed during backward movement are alternately arranged in each raster and each column. It is. In the first embodiment, printing is performed using the slashed dot arrangement shown in FIG. With such a dot arrangement, the print head 90 performs color printing with multicolor dots, for example, with a resolution of 720 × 720 dpi. Note that the resolution of printing by the print head 90 may be 180 × 180 dpi or higher, for example, 180, 240, 360, 720, 1440 dpi or higher in the main scanning direction and 180, 240, The resolution may be appropriately combined with a resolution of 360, 720, 1440 dpi, or higher. Alternatively, it may be a resolution of 300 or 600 dpi (or higher). Since a method for realizing a desired dot arrangement using interlace or overlap at various resolutions is well known, detailed description thereof is omitted.

  The control unit 30 that executes printing by controlling the print head 90, the carriage motor 70, the paper feed motor 74, and the like is configured by connecting a CPU 40, a ROM 51, a RAM 52, and an EEPROM 60 to each other via a bus. The control unit 30 develops a program stored in the ROM 51 or the EEPROM 60 in the RAM 52 and executes it to control the overall operation of the printer 20 and also functions as a dot data generation unit 42 and a printing unit 43 in claims. . Details of these functional units will be described later.

  The EEPROM 60 stores a first dither mask 61 and a second dither mask 62. The first dither mask 61 and the second dither mask 62 used in this embodiment have a size of 64 × 64, and threshold values from 0 to 256 are stored in 4096 storage elements. Yes. Each threshold value is used in halftone processing described later. The arrangement of each threshold value as the second dither mask 62 is determined so as to have a characteristic close to a so-called blue noise mask. The characteristics of the second dither mask 62 used in the present embodiment will be described in detail later. As in the case of the first dither mask 61 that realizes high image quality, the second dither mask 62 is configured as a highly dispersed dispersive dither mask. Yes.

  In this embodiment, printing is performed by the printer 20 alone. A memory card slot 98 is connected to the control unit 30, and image data ORG is read from the memory card MC inserted into the memory card slot 98 and inputted. In this embodiment, the image data ORG input from the memory card MC is data composed of three color components of red (R), green (G), and blue (B). The printer 20 performs printing using the image ORG in the memory card MC. Of course, it is possible to adopt a configuration in which a computer is connected to an external computer using a USB or a LAN, halftone processing is performed on the computer side, and the result is received and printed by the printer 20.

  The printer 20 having the above hardware configuration drives the carriage motor 70 to reciprocate the print head 90 in the main scanning direction with respect to the printing paper P, and drives the paper feed motor 74. Thus, the printing paper P is moved in the sub-scanning direction. The control unit 30 performs printing by driving the nozzles at an appropriate timing based on the print data in accordance with the movement of the carriage 80 in the reciprocating motion (main scanning) and the paper feeding movement of the printing medium (sub scanning). Ink dots of appropriate colors are formed at appropriate positions on the paper P. By doing so, the printer 20 can print the color image input from the memory card MC on the printing paper P.

A-2. Printing process:
A printing process in the printer 20 will be described. FIG. 4 is a flowchart showing the flow of printing processing in the printer 20. The printing process here is started when the user performs an instruction to print a predetermined image stored in the memory card MC using the operation panel 99 or the like. When the printing process is started, the CPU 40 first reads and inputs RGB format image data ORG to be printed from the memory card MC via the memory card slot 98 (step S110).

  When the image data ORG is input, the CPU 40 performs a process of accepting setting of paper information through the operation panel 99 (step S115). This paper information includes at least a paper size and a paper type. Specific examples of the paper type (media type) include “plain paper”, “light paper”, “thick paper”, and the like, and are information on the mass per unit area of the paper. From the paper type, the paper thickness can be estimated. Note that a predetermined paper size (for example, A4) and a paper type (for example, plain paper) are set by default, and default paper information is used unless otherwise specified by the user.

  After accepting the setting of the paper information, the CPU 40 refers to a lookup table (not shown) stored in the EEPROM 60 and color-converts the RGB format of the image data ORG to the CMYKLcLm format (step S120).

  When the color conversion process is performed, the CPU 40 determines whether the print mode is set to the high image quality mode or the high speed mode (step S125). If it is determined that the high-quality mode is set (step S125, high-quality mode), the first halftone process (step S130) is performed, and if it is determined that the high-speed mode is set (step S125, high-speed mode). Mode) and second halftone processing (step S135). In the high-speed mode, the dot formation position shifts more easily than in the high-quality mode. Therefore, in the case of the high-speed mode, the second halftone process is performed in which the image quality hardly deteriorates even if the dot formation position shifts. On the other hand, in the high image quality mode, the dot formation position is unlikely to shift. Therefore, when dots are formed according to the dot data, the first halftone process is performed to achieve higher image quality. In both the first and second halftone processes, the dot data generation unit 42 performs a process of converting image data into ON / OFF data (hereinafter also referred to as dot data) of dots of each color. This processing is performed using a dither method in this embodiment. That is, the input data is compared with the threshold stored in the storage element at the position corresponding to the input data among the plurality of thresholds constituting the dither mask, and if the input data is larger than the threshold, a dot is formed. If (dot ON) is determined and the input data is equal to or less than the threshold value, it is determined that no dot is formed (dot OFF). The first dither mask 61 is used in the first halftone process, and the first dither mask 62 is used in the second halftone process. The first dither mask 61 or the second dither mask 62 used in this process is repeatedly applied in the main scanning direction and the sub-scanning direction to each input data arranged in the main scanning direction and the sub-scanning direction. The halftone process in the present embodiment is controlled so that the generated dot data has a predetermined characteristic. The contents of this control depend on the properties of the first dither mask 61 and the second dither mask 62. The characteristics of the first dither mask 61 and the second dither mask 62 will be described later. The halftone process is not limited to the binarization process of dot ON / OFF, but may be a multi-value process such as ON / OFF of large dots and small dots. Further, the image data provided to step S130 or step S135 may be subjected to image processing such as resolution conversion processing or smoothing processing.

  When the first or second halftone processing is performed, the CPU 40 performs overlap and interlace processing for rearranging the dot pattern data to be formed in one main scanning unit in accordance with the nozzle arrangement of the printer 20 and the paper feed amount. Is performed (step S140). When the overlap and interlace processing is performed, the CPU 40 performs printing by driving the print head 90, the carriage motor 70, the motor 74, and the like as processing of the printing unit 43 (step S150).

  The arrangement of dots formed by such printing processing will be described. As is clear from the above description, the printer 20 has a print head at a plurality of different timings (ie, forward and backward movements) while changing the ink ejection position with respect to the print medium in a common print area of the print medium. The ink is ejected from the ink to form dots, and the dot formed by forward movement (hereinafter also referred to as forward movement dot) and the dot formed by backward movement (hereinafter also referred to as backward movement dot) are combined with each other The printed image is output. In the first embodiment, since the dot arrangement is a task arrangement (FIG. 3C), the dots formed when the print head 90 moves forward are hatched in FIG. 5A. The dots formed at the checkered pixel positions when the print head 90 moves backward are shaded in the column direction as shown by the shaded area in FIG. 5B. It is formed at pixel positions of a staggered arrangement shifted by one pixel. A set of pixels corresponding to dots formed during forward movement is referred to as a first pixel group, and a set of pixels corresponding to dots formed during backward movement is referred to as a second pixel group. In FIGS. 5A and 5B, the actually formed dots are indicated by “●” marks and hatched “◯” marks, respectively. The dot size is usually set larger than the diagonal size of the pixel so that the print medium surface can be 100% covered with the maximum density even if the dot formation position is slightly shifted. As shown in FIG. 5C, the printed image is a combination of dots formed in each of the first and second pixel groups.

  If the formation positions of the dots formed during the forward movement and the backward movement are completely adjusted, as shown schematically in FIG. 3, the dots are completely formed in both the main scanning direction and the sub scanning direction. In alternate positions. This state is called that the inter-dot distance is zero. FIG. 5C shows a state in which the distance between dots is 0, that is, the formed forward movement dots and backward movement dots are completely formed alternately. Since the position where dots are formed in bi-directional printing may be shifted and printing may be performed in a state different from the state where the inter-dot distance is 0, the dots actually formed are as shown in FIG. May be different. For example, assuming that the dot formation position during forward movement is shifted by about one pixel in the raster direction (main scanning direction) with respect to the dot formation position during backward movement, in the example shown in FIG. As shown in FIG. 6A, as a result of the dots formed during forward movement being displaced in the main scanning direction, the overlapping area of the dots increases. Further, as shown in FIG. 6B, when the deviation amount increases to 2, the overlapping area further increases.

  As can be seen from FIG. 5C, there is very little overlap between dots if there is no misalignment. This is because the dither mask having the blue noise characteristic tries to arrange the dots as far apart as possible. On the other hand, in actual printing, when a positional deviation occurs in the dot formation position and the inter-dot distance increases, dots are formed during forward movement as shown in FIGS. 6 (A) and 6 (B). The overlapping amount of the dots belonging to the first pixel group and the dots belonging to the second pixel group in which dots are formed during the backward movement increases. As the dot overlap amount increases, the coverage, which is the ratio of dots covering the printing paper P, varies. In addition, if there is no deviation in the dot formation position, it is possible that a dot that was not adjacent to one another may be formed in an adjacent position due to the increase in the inter-dot distance and the deviation of the dot formation position. In this case, the coverage varies for the portion corresponding to the overlap due to the dot gain. In this embodiment, since the resolution is 720 × 720 dpi, a shift of 2 pixels means that the dot formed with the pixel group to which the dot group formed at the time of forward movement belongs to the pixel group to which the dot group at the time of backward movement belongs. The distance between the two corresponds to 2/720 inch. Even if the same two pixels are displaced, if the resolution is, for example, 300 × 300 dpi, the distance between dots is 2/300 inches, which is a large value.

A-3. Halftone processing:
Based on the above points, the characteristics of the halftone process in the first embodiment will be described. In the halftone process shown as step S130 in FIG. 4, the gradation values of the pixels belonging to the first pixel group and the pixels belonging to the second pixel group are compared with the second dither mask 62 stored in the EEPROM 60. Thus, it is determined whether or not to form a dot at each pixel position. On the other hand, in the halftone process shown as step S <b> 135, the gradation values of the pixels belonging to the first pixel group and the pixels belonging to the second pixel group are compared with the second dither mask 62 stored in the EEPROM 60. Thus, it is determined whether or not to form a dot at each pixel position. Data indicating ON / OFF of the determined dot is referred to as dot data.

  As described above, the second dither mask 62 used to generate the dot data is set to be highly dispersive although it is lower than the blue noise mask. “High dispersibility” means that the distribution of dots formed has a noise characteristic having a peak in a higher frequency side than a low frequency region below a predetermined spatial frequency in the spatial frequency region. The arrangement of the threshold values in the second dither mask 62 in the first embodiment is such that, for the dot arrangement in which the dots formed in each of the forward movement and the backward movement are combined, the ratio of the dots overlapping or adjacent to each other is as follows: For example, it is determined as being lower than the dot arrangement according to the white noise characteristic but higher than the dot arrangement using the blue noise mask. For this reason, when the second dither mask 62 of the first embodiment is used, even if the dot formation position shifts in the reciprocating motion, the variation in the coverage is suppressed.

Moreover, in the second dither mask 62 used in the halftone process of the present embodiment, the dot-to-dot distance formed by the forward and backward movements is an arbitrary distance in the range of 2/720 inches to 5/720 inches. Even if printing is performed, the threshold values of the second dither mask 62 are arranged so that the change in the CIEL * a * b * color space of the formed image falls within the following range. Is set.
(A) When the image to be printed is yellow (RGB value, R = G = 255, B = 0-64, where RGB value is converted into 8 bits), it is in the CIEL * a * b * color space. The change in b * is in the range of 2 or less.
(B) When the image to be printed is magenta (RGB value, R = B = 255, G = 0-64, where RGB value is converted into 8 bits), it is in the CIEL * a * b * color space. The change of L * is in the range of 0.5 or less or the change of a * is in the range of 0.5 or less.
(C) When the image to be printed is cyan (RGB value, G = B = 255, R = 0-64, where RGB value is converted to 8 bits), it is in the CIE L * a * b * color space. The change in L * is in the range of 0.5 or less.

  A plurality of dot-to-dot distance differences from the position where the dots belonging to the first pixel group are formed to the position where the dots belonging to the second pixel group are formed are 2/720 inches to 5/720 inches. The state means that the distance between one dot in a plurality of states where printing is performed and the distance between dots in the other state are different by 2/720 to 5/720 inches. In general, the forward and backward dot formation positions in bi-directional printing are adjusted by some method. Therefore, the adjusted state is referred to as a first state (first setting). The state after the deviation has occurred is referred to as a second state (second setting), and may be handled as a plurality of states having different inter-dot distances. Assuming that this is the case, the first state is that the dots belonging to the first pixel group have a default position (a distance between dots) that is not shifted from the formation position of the dots belonging to the second pixel group. (Distance = 0 position), that is, the dot belonging to the first pixel group can be determined as being in a position to be formed with respect to the dots belonging to the second pixel group. Alternatively, the dots belonging to the first pixel group may be in a predetermined position in relation to the formation positions of the dots belonging to the second pixel group. In contrast to the first state, the second state is a dot at a position shifted by an arbitrary distance in the range of 2/720 inch to 5/720 inch from the first position where the dot is formed in the first state. Is formed. In addition, the first and second states may be defined with the first state as a default state (reference state) at the time of shipment. As a result of the halftone process using the second dither mask 62 having the above characteristics when forming dots in each of the first and second states, the printer 20 of this embodiment has the following effects. Play.

A-4. Effects of the embodiment:
In the printer 20 of the first embodiment having the above configuration, the image data ORG is received, and the control unit 30 performs the processing shown in FIG. At this time, the second dither mask 62 is used for the gradation value of each pixel constituting the image in the high-speed mode, that is, when the reciprocating speed of the carriage 80 is likely to cause a shift in dot formation position. Halftone processing is performed by the dither method, and the dot distribution is converted. Since the second dither mask 62 is originally a dispersion type dither mask typified by a blue noise mask giving priority to dispersibility, the image quality of an image expressed by dot distribution after halftone processing is granular. The feeling is low and the reproducibility of the image is high.

  In addition, even when printing is performed in a plurality of states where the dot formation positions in the reciprocating movement are different, fluctuations in the overlapping of the dots formed during the forward movement and the dots formed during the backward movement are suppressed, and the formed image The change in the CIEL * a * b * color space is within the range from (A) to (C). In order to realize such characteristics, in the first embodiment, the pixels of the first pixel group to which the dots formed during the forward movement of the print head 90 belong and the second pixel group to which the dots formed during the backward movement belong. The probability that a dot is formed in a pixel adjacent to the other pixel, that is, a pair pixel, is set higher than that of the blue noise mask. As a result, even if there is a deviation in the dot formation position during forward movement and backward movement, the variation in coverage due to the formed dots is small, and the gradation value and color of the image are less likely to change. Have. This point will be described with reference to FIGS.

  FIG. 7 shows a dot formation position deviation amount (pixel unit) and a change amount ΔL of L * in the L * a * b * color space when a uniform cyan image is formed with a resolution of 720 × 720 dpi. It is explanatory drawing which shows a relationship. FIG. 7A shows an example of L when G = B = 255 and R is changed to 0, 32, 64, 96, and 128 when the color of an image to be printed is expressed in RGB (each color is 8 bits). The value of * is shown. In FIG. 7A, P is a value of L * when the second dither mask 62 of the present embodiment is used, and N is a value of L * when the conventional blue noise mask is used. The values measured in accordance with the deviation amount of the dot formation position (−10 pixels, −5 pixels, 0 pixels (no deviation), 2 pixels, 5 pixels, 10 pixels, 20 pixels) are shown. The change amount ΔL of each L * is shown at the right end of FIG. The change amount ΔL is the difference between the maximum value and the minimum value among L * for each dot formation position shift amount.

  Of the data shown in FIG. 7A, a graph plotting the amount of deviation of the dot formation position on the horizontal axis with the values of P and N in the case of R = 0 as representative examples is shown in FIG. It is. Each data shown in FIG. 7A can be similarly graphed. As can be seen from FIG. 7B, in the halftone process using the second dither mask 62 of this embodiment, the dot formation position is positive from the state where there is no deviation during forward movement and backward movement.・ It can be seen that L * hardly changes even if it deviates in either direction. The change amount ΔL is suppressed to 0.3 (in the case of R = 32, 64, 96) at the maximum. On the other hand, in the halftone process using the conventional blue noise mask, if there is a deviation in the dot formation position between the forward movement and the backward movement, in the plus or minus direction from the state where there is no deviation. It can be seen that L * changes greatly even if the dot formation position is shifted. The change amount ΔL increases to 1.4 at the maximum (when R = 0). In general, if the change in L * exceeds the range of 0.5, it can be visually recognized as a difference. Therefore, by suppressing the variation in the coverage to a range corresponding to this or less, Even if there is a deviation (difference in the distance between dots) between the position of the dot formed during the forward movement and the position of the dot formed during the backward movement, it is not known as a change in brightness.

  Similarly, as shown in FIGS. 8A and 8B and FIGS. 9A and 9B, even when a magenta or yellow uniform image is formed, the change in the CIEL * a * b * color space. The range of is sufficiently small. For example, as shown in FIG. 8A, in magenta (R = B = 255, G = 0, 32, 64, 96, 128), in this embodiment, the change in L * is up to 0.4. The variation of the coverage is suppressed so as to be in the range. On the other hand, when the blue noise mask is used, the change ΔL in L * is 1.5 at the maximum (when G = 32). Further, as shown in FIG. 8B, for yellow (R = G = 255, B = 0, 32, 64, 96, 128), the change ΔL of L * in this embodiment is 1.3. While the variation of the coverage is suppressed so as to be in the range, when the blue noise mask is used, it is 5.7 (when B = 64).

  Note that the change in the image due to the change in the coverage is not limited to L * corresponding to the brightness. As shown in FIGS. 9A and 9B, when the second dither mask 62 in the present embodiment is used, cyan (G = B = 255, R = 0, 32, 64, 96, 128), the variation in the coverage ratio is suppressed within a range where the change Δa of a * does not exceed 0.6. On the other hand, in the conventional blue noise mask, the change Δa of a * is up to 1.6 (when R = 96). Further, in magenta (R = B = 255, G = 0, 32, 64, 96, 128), in this embodiment, the variation Δb of b * is suppressed within a range not exceeding 0.4. ing. On the other hand, when a blue noise mask is used, the change Δb in b * is 1.6 at the maximum (when G = 96). Such a color change is caused by a difference in the formation ratio of each color ink droplet to be formed. In this embodiment, printing is performed using ink droplets of CMYLcLm, and the formation ratio of each color ink droplet varies depending on the color of the image. As a result, if the variation in the coverage due to the deviation of the dot formation position is different for each ink color, the color will also change. In this embodiment, the variation in the coverage due to the deviation of the dot formation position (difference in the distance between dots) is suppressed so that the change in a * and b * falls within a predetermined range. It is never perceived as a change in color. In particular, one of a plurality of states is a state in which the positions of dots formed at the time of forward movement and the positions of dots formed at the time of backward movement are printed at a sufficiently adjusted reference position, and the other one is Even when printing is performed with an arbitrary distance in the range of 2/720 inch to 5/720 inch from this reference position, the range of change in the L * a * b * color space is sufficiently small and suppressed. It is not recognized as a change in brightness or color.

  As a result, according to the halftone processing of the present embodiment, even if a deviation (difference in the distance between dots) occurs in the dot formation position during forward movement and backward movement, the gradation value of the image to be formed Little change occurs. Moreover, it hardly occurs that the color changes with the variation of the coverage. The color change in the first embodiment is suppressed to at least the range visually recognized by the user.

  On the other hand, in the case of the high image quality mode, that is, when the reciprocating speed of the carriage 80 is low and thus it is difficult for the dot formation position to shift, the first dither mask 61 is used and the second dither mask 62 is used. Therefore, the graininess can be improved. Therefore, a high-quality printed image can be obtained.

B. Dither mask generation method:
In this embodiment, when the formation position of the dots formed by the reciprocating movement is shifted, the variation in the coverage is suppressed to a predetermined range in the CIEL * a * b * color space. In this embodiment, the suppression of the variation in the coverage is realized by controlling the probability of occurrence of paired dots. Hereinafter, the principle of controlling the probability of occurrence of such a pair of dots will be described, and then a method for generating a dither mask that controls the probability of occurrence of such a pair of dots will be described.

B-1. Principle of paired dot control:
In the dot-concentrated dot arrangement, the variation in coverage that occurs when the dot formation position in the reciprocating movement is deviated has already been described with reference to FIG. In the case of the dot concentration type, the coverage generally tends to increase as the dot formation position shifts, whereas when the dot dispersion type dither mask is used, the coverage increases as the dot formation position shifts. The rate tends to decrease. For this reason, when a blue noise mask, which is a typical dot-dispersed dither mask, is used, the brightness of the image generally increases if there is a shift between the dot formation positions in the forward and backward movements. Sometimes the color changes. On the other hand, in the first embodiment, the dot formed during the reciprocating movement and the dot formed during the forward movement when the dot is shifted by an arbitrary distance in the range of 2/720 inch to 5/720 inch and the backward movement are used. Variations in the way of overlapping with the dots to be formed are suppressed so that changes in the CIEL * a * b * color space of the formed image are within a predetermined range. In addition to this, the dot formation position shift may occur for each nozzle. For example, there are a difference in the formation position of the nozzle head, a difference in the landing position of the ink droplet due to the extension of the print medium, and the like.

Various methods can be considered as a method for giving the above-mentioned characteristics to the dither mask used in the halftone process. In the first embodiment, in order to give the second dither mask 62 the above characteristics, a pair dot control method described below is used. A pair dot means a relationship between both dots when a dot formed at the time of forward movement (reverse movement) and a dot formed at the time of backward movement (forward movement) are arranged at adjacent positions. In the second dither mask 62 of the first embodiment, as will be described later, the arrangement of threshold values is set so that pair dots are generated with a significant probability even in a low density region. The significant probability that a pair dot occurs is a probability set as follows. In the second dither mask 62 used in the first embodiment, when the gradation value of the image data is in the range of 0 to 128/255, dots are arranged in each pixel belonging to the first and second pixel groups. Where k (0 ≦ k ≦ 1) is a pixel adjacent to the right in the raster direction (main scanning direction) or a pixel adjacent in the column direction (sub-scanning direction) of one pixel in which dots are formed. The probability K of forming crab dots is about 0.8 × k ² , respectively.

  Of the pixels adjacent to one pixel of interest, those in which the group in which the dots are formed are different are hereinafter referred to as “adjacent pixels”. There are four pixels adjacent to the pixel of interest in the vertical and horizontal directions in the layout shown in FIG. 3C. A large shift in the dot formation position occurs between the dot formed during forward movement and the dot formed during backward movement. Therefore, not only whether or not they are adjacent to each other, but also the dot generation probability is adjusted only for pixels that are adjacent and belong to different pixel groups. In the first embodiment, the dots formed at the time of forward movement and the dots formed at the time of backward movement are staggered as shown in FIG. Pixels belonging to the pixel group are present at four locations on the top, bottom, left and right of the pixel of interest. In this embodiment, only the pixels adjacent to the right in the raster direction (main scanning direction) and the column direction (sub-scanning direction) are set as “adjacent pixels” for the pixel of interest. This is because it is only necessary to count the paired dots (dots formed on both of the adjacent pixels) in consideration of only one of the adjacent pixels that are point-symmetric with respect to the target pixel. For all the pixels forming the image, if you count only one of the adjacent pixels that are point-symmetric while sequentially moving the pixel of interest from the upper left to the lower right of the image, all the pairs will not overlap. Can count dots.

  In FIG. 10A, when the position of the pixel of interest OJ is (0, 0) and the right direction of the main scanning and the downward direction of the sub scanning are positive, the position (1, 0) is the adjacent pixel NR on the right side and the position ( 0, 1) indicates that it is the lower adjacent pixel ND. When specifying the relationship between the pixel of interest OJ and any one of the adjacent pixels NR and ND, these are collectively referred to as “pair pixels”. In the first embodiment, as described above, the adjacent pixels constituting the pair pixel together with the target pixel are limited to the pixels NR and ND on the right or lower side of the target pixel OJ. The number of paired dots can be counted only for the pixels. In FIG. 10A, the pair pixel is limited to the pixel adjacent to the pixel of interest, but the pixel that considers the occurrence probability as the pair pixel is not necessarily limited to the adjacent pixel. As shown in FIGS. 10B and 10C, even a pixel separated from the target pixel can be treated as an adjacent pixel, and another example of such a case will be described later.

The probability that a dot is formed in a pair pixel will be described. Here, the gradation value is treated as corresponding to the probability that the dot is turned on (formed). Assuming that the image ORG to be halftone processed is a uniform image having a gradation value of 26/255, the dot arrangement is about one in 10 pixels (k = 0.1). On the other hand, the probability K of forming dots in the paired pixels is about K = 0.8 × k ² ≈0.008 in the dither mask of the first embodiment. In the conventional dither mask with high dispersibility, priority is given to the dispersibility of dots in the low density region, and dots are formed in paired pixels which are adjacent pixels, and the probability is as close to 0 as possible. In fact, in a dither mask having characteristics known as a blue noise mask, no example was found in which dots were formed on paired pixels with a gradation value of 26/255.

On the other hand, in the first embodiment, when the gradation value is 0 to 127/255, that is, the dot formation probability k is in the range of about 0 to 0.5, the probability K that the dots are formed together in the paired pixels is It is about 0.8 × k ² . That is, for example, if the gradation value is 52/255 (k≈0.2), the probability K that dots are formed on the paired pixels is 0.032, that is, about 3 sets per 100 paired pixels. Dots are formed at a rate.

  FIG. 11 schematically shows the rate at which dots are formed in a pair of pixels. In FIG. 11, the horizontal axis represents the probability that a dot is formed in a pixel and corresponds to the gradation value of the image. In addition, the vertical axis in FIG. 11 indicates the rate at which dots are formed on the paired pixels. In FIG. 11, a solid line JD1 indicates a case where halftone processing is performed using the dither mask of the present embodiment, and an alternate long and short dash line BN1 indicates a case where halftone processing is performed using a blue noise mask. Yes. A broken line WN1 indicates a case where halftone processing is performed using a white noise mask. Here, the white noise mask is set so that the mask size is sufficiently large, and each threshold value of the dither mask is set with a random number so that a result equivalent to the random dither method in which the threshold value is generated with a random number every time can be obtained. To the dither mask. The blue noise mask exhibits a blue noise characteristic that does not include a low frequency component, whereas the white noise mask exhibits a white noise characteristic that uniformly includes a low frequency component to a high frequency component.

As shown in the figure, when the blue noise mask is used, in the region where the gradation value of the image is low (gradation value 0 to 51, dot generation probability k = 0 to 0.2), both of the pair pixels are dotted. The probability of forming is almost zero. On the other hand, when the white noise mask is used, since the dot formation position is random, the probability that a dot is formed on a pair of pixels is substantially equal to k ² with respect to the dot formation probability k. ing. With respect to these characteristics, the dither mask adopted in this embodiment is a distributed dither mask, but the probability K that dots are formed on the paired pixels is the gradation value as shown by the solid line JD1. In the range of 0 to 127 (dot generation probability k = 0 to 0.5), it is approximately 0.8 × k ² . In other words, the dither mask used in this example shows the dispersibility close to that of the blue noise mask for the distribution of dots to be formed, and the white noise mask for the probability K that both dots are formed on the paired pixels. It shows close characteristics. A method of creating a distributed dither mask with an increased dot formation rate in such a pair of pixels will be described later.

  FIG. 12 is a graph showing the simulation result of the variation in coverage when the image data ORG having the gradation value with the dot formation ratio of 96/255 is processed. In the figure, the horizontal axis shows the amount of deviation of the dot formation position during forward movement and backward movement in units of pixels, and the vertical axis shows the coverage rate variation rate. In the graph of FIG. 12, the solid line JE1 is the case where the second dither mask 62 of the first embodiment is used, and the broken line BB1 is a typical blue noise mask created so that dots are generated as discretely as possible. The case where is used is shown respectively. Here, the coverage means the ratio of the formed dots covering the paper P, and the fluctuation of the coverage is the ratio of the dots covering the printing paper P when the original dot formation position is not displaced. As a reference, it means the degree to which the dot formation position has changed, so that the dot overlap occurs and the ratio of covering the paper changes.

  In the printer 20, the dot size is set slightly larger than the pixel size so that the ink droplets completely cover the pixel. For this reason, even if the dots do not overlap, if the dots are in contact with each other, dot overlap occurs, and the coverage is reduced. Since a typical blue noise mask tries to disperse the dots as far as possible from each other, contact between dots without any deviation, that is, a factor of lowering the coverage rate is minimized. Accordingly, when a deviation occurs in the dot formation position in, for example, reciprocal printing in the actual printer 20, the dot formation position is lost from the optimal arrangement, and the contact and overlap between the dots increase, and the coverage rate generally decreases. When the data of the same gradation value is printed, if the coverage ratio fluctuates, the image becomes uneven and the image quality deteriorates. Such unevenness in image quality due to the variation in coverage is particularly noticeable in large-format printed matter printed by a large-format printer. This is because large-sized printed materials are usually viewed from a distance, and when viewed from a distance, low-frequency printing unevenness is easily noticed.

  As shown in FIG. 12, when the second dither mask 62 of the present embodiment is used, even if the dot formation position is deviated between forward movement and backward movement, a normal distributed dither mask is used. It can be seen that the variation in the coverage is less likely to occur than when it is used. In FIG. 12, when the shift amount Δd of the reverse printing position with respect to the forward printing position is an odd number (Δd = 1, 3,...) When the number of pixels is an even number (Δd = 2, 4,...). ..) The rate of decrease in the coverage rate increases, and fluctuations with a deviation amount of 2 as a cycle are observed. The variation of the deviation amount in two cycles occurs in printing by the layout shown in FIG. 3C, when the horizontal deviation amount is an odd number, dots formed during forward movement and dots formed during backward movement. This is because the positions of are completely overlapped. For this reason, in the simulation executed on the assumption that all the dot formation positions are shifted in the same manner by the reciprocation without considering the shift of the dot formation positions due to other factors, when the horizontal shift amount is an odd number, FIG. As shown in the above, a decrease in the coverage rate becomes obvious. In the actual printer 20, since a small positional shift in units of pixels is superimposed on the positional shift of dots in reciprocating printing, the variation in coverage shown in FIG. 12 is flattened. The variation in the coverage ratio when the second dither mask 62 of the present embodiment is used becomes even more flat at the solid line JF1 in FIG. On the other hand, the fluctuation when the distributed dither mask is used is flattened somewhat from the broken line BB1 in FIG. 12, but cannot be completely eliminated, and the fluctuation of the coverage remains.

The reason why the paired dot formation ratio is made closer to k ² is based on the following knowledge. In other words, even if the dot spacing is dispersed and arranged as far as possible using a blue noise mask, etc., if the dot formation position of a specific pixel group shifts and the shift amount becomes sufficiently large, dots in a specific direction are adjacent to each other. Thus, the knowledge that the probability of pair dots converges to k ² was obtained. As a result of investigating the actual blue noise mask, as shown in FIG. 13, it was found that the occurrence rate of paired dots converges to a substantially constant value k ² when the shift amount is 4 to 5 pixels or more. This is because, when the shift amount is large, two pixels that are originally separated from each other are adjacent to each other. If the distance between the two pixels is sufficiently large, the correlation between the presence or absence of dot formation in both pixels decreases, so the probability that dots are simultaneously formed in both pixels is simply the tone value of both (the probability of dot formation) k) is multiplied by k ² . Therefore, if the pair dot occurrence rate in a state where there is no deviation is made close to k ² in advance, the pair dot occurrence rate does not change much when any deviation occurs, and the variation in coverage can be suppressed.

In the first embodiment, the probability K that both dots are formed on the paired pixels is
K = 0.8 × k ²
It is said. Accordingly, it is possible to suppress a decrease in the dispersibility of the dot distribution when there is no deviation in the dot formation position. This coefficient adjusts the probability of occurrence of pair dots. If the coefficient is 0.8, it means that the occurrence rate of pair dots is suppressed to 80%. Of course, the coefficient can be appropriately set as long as it is in the range of, for example, about 0.6 to 1.4. If the coefficient is in the range of 0.8 to 1.2, the variation in the probability of occurrence of paired dots due to the shift in the dot formation position can be suitably suppressed. This is desirable from the viewpoint of suppressing fluctuations in the occurrence probability. Further, if priority is given to the dispersibility of dots in the low gradation region, the coefficient can be adjusted to a value of 0.8 or less, for example, about 0.6 to 0.8.

B-2. Dither mask generation method:
The dither mask used in the first embodiment described above was generated by the following method. FIG. 14 is a flowchart illustrating an example of a dither mask generation method used in the first embodiment. In this embodiment, prepared blue noise mask, from the blue noise mask to produce a dither mask close the probability that both dots are formed in the pair of pixels in k ^2. The generated dither mask is hereinafter referred to as a “pair pixel control mask”. The dither mask being created is referred to as a “working mask”.

  When generating a pair pixel control mask, a blue noise mask is first prepared (step S200). In this example, a 64 × 64 blue noise mask was used. In the blue noise mask of this example, 255 threshold values from 0 to 254 are stored in a matrix having a size of 64 × 64. Next, a process of counting the number of pair dots for each gradation value over the entire gradation range is performed for the current working mask (step S210). Specifically, this process is a process of counting the right adjacent pair dot number RPD [1, 2,... 127] and the lower adjacent pair dot number UPD [1, 2,. is there. In the following description, when parentheses are used as in (S), the value at the gradation value S is indicated, and when [] is used as in [a,. It shall represent the arrangement from a to x. In addition, the arrangement from the gradation range a to x is expressed as [a: x].

  Since all the threshold values are known for the working mask, the dot formation position at each gradation value can be examined in the gradation value range of 1 to 127/255. For this reason, it is easy to count the right adjacent pair dot number RPD (S) and the lower adjacent pair dot number UPD (S) for each gradation value S. Here, the count of the number of paired dots is limited to the gradation values of 1 to 127/255. The paired pixel control mask used in the first embodiment, that is, the paired dots in the gradation range of 1 to 127/255. This is to generate a mask having a predetermined characteristic of the occurrence probability. As the gradation value S increases, the number of pair dots also approaches the probability of occurrence here in the blue noise mask, so instead of counting the number of adjacent pair dots for the entire range, gradation values 1 to 127/255 In this range, the probability of occurrence of paired dots may be adjusted. Of course, the method described below is also applicable to the case where the number of pair dots is counted and the probability of occurrence is adjusted for the entire gradation range.

In step S210, after counting the right adjacent pair dot number RPD [1: 127] and the lower adjacent pair dot number UPD [1: 127] in a predetermined gradation range (here, 1 to 127/255), each gradation value is counted. It is determined whether the number of paired dots for each S is within the target range M (S) (step S220). Here, the target range M (S) is a range set as follows. If the dither mask has white noise characteristics, the dots are generated randomly. If the probability that a dot is formed in one pixel is k, the right or lower adjacent pixel The probability that dots are formed (this is referred to as the probability of occurrence of paired dots) is k ² respectively. When the gradation value of the image is 1,
k = 0.00392156 (= 1/255)
And the probability of occurrence of paired dots is
k ² = 0.0000154
It becomes. Therefore, when it is assumed that dots are randomly formed, a value H (hereinafter referred to as a predicted value) H that is predicted to have a pair pixel in 64 × 64 pixels is:
H = k ² × 4096 = 0.063 ≒ 0
It is. This calculation is repeated in advance in the range of gradation values 1 to 127/255 to obtain a theoretical predicted value H [1: 127] of a paired dot, and this is multiplied by a coefficient of 0.8 for each gradation. The target value m [1: 127] of the paired dot in the value S is obtained in advance. In the present embodiment, the target value m (S) has a width of ± 20%, and this is referred to as a target range M (S).

  FIG. 15 shows paired dot predicted values H [1:32] and target values m [1:32] when the gradation value S is 1 to 32. In FIG. As shown in the figure, the gradation value S = 10, the prediction value H (10) = 6, the target value m (10) = 5, the gradation value S = 20, the prediction value H (20) = 25, the target value. It can be seen that m (20) = 20.

  In step S220, the theoretical pair dot target range M [1: 127] thus obtained is compared with the right adjacent pair dot number RPD [1: 127] and the lower adjacent pair dot number UPD [1: 127]. To do. As a result of comparison, if it cannot be determined that both the paired dot numbers RPD [1: 127] and URD [1: 127] are within the target range M [1: 127], then the threshold value in the work mask is set. Among them, an appropriate number of threshold values (for example, two threshold values) are randomly replaced (step S230). Since the pixel groups are randomly switched, the thresholds corresponding to the same pixel group may be interchanged, or may be interchanged between different pixel groups.

  When the threshold value in the work mask is changed, the number of pair dots in each gradation value changes, so that the number of pair dots is corrected by changing the threshold value (step S240). Since the number of paired dots changes only within the range of gradation values corresponding to the replaced threshold value, the number of paired dots is not counted again in the gradation range of 1 to 127/255. For example, the threshold value p and the threshold value q ( If p <q) is exchanged, only the right adjacent pair dot number RPD [p: q] and the lower adjacent pair dot number UPD [p: q] may be recounted. Although the threshold value to be replaced is selected at random, since the generation characteristics of the paired dots are to be adjusted in the range of gradation values 1 to 127/255, at least one of the threshold values to be replaced is within this range. It is desirable to use a threshold value.

The number of pair dots thus counted is checked, and then it is determined whether or not the pair dot characteristics have been improved (step S250). Here, whether or not the paired dot characteristics are improved is determined as follows.
(A) If the right and lower adjacent pair dot numbers RPD [p: q] and UPD [p: q] are close to k ² due to the replacement of the threshold values, it is determined that they have been improved.
(B) When one of the right and lower adjacent pair dot numbers UPD [p: q] and [p: q] approaches k ² and the other has not changed due to the replacement of the threshold value, it is determined that the improvement has occurred. .
(C) When the gradation range [p: q] is improved or partially deteriorated, the number of pair dots generated in each gradation value of the gradation range and the predicted value at the gradation value If the sum of the differences is small, it is judged as an improvement.

If the above determination is made and it is determined that the paired dot characteristics are not improved, the process returns to step S230, and the process is executed again from the process of randomly changing the threshold value. If the replacement of the threshold is to replace two thresholds, the combination is the entire range of gradations,
There will be ₄₀₉₆ C ₂ ways. Even if it is limited to the range of gradation value 1 to 127/255,
_{There are 2048} C ₂ ways. Accordingly, a considerable number of combinations of thresholds are replaced, and although it takes a considerable amount of time to complete all cases, replacement is found to improve the paired dot characteristics if performed sequentially (step S250, “YES”). .

  Therefore, if it is determined that the paired dot characteristics have been improved, it is next determined whether or not the graininess characteristics are satisfactory (step S260). Here, that there is no problem with the graininess characteristic means that the graininess index shown below is within the target range, or is not within the target range, but is improved from before the threshold replacement. Since the granularity index is a known technique (for example, Japanese Patent Application Laid-Open No. 2007-15359), a detailed description is omitted, but the power spectrum FS is obtained by Fourier transforming the image to obtain the power spectrum FS, This is an index obtained by adding a weight corresponding to a sensitivity characteristic VTF (Visual Transfer Function) with respect to a visual spatial frequency possessed by a human and integrating at each spatial frequency. FIG. 16 shows an example of VTF. Various formulas have been proposed as empirical formulas that give such VTF, and the following formula (1) shows typical empirical formulas. The variable L represents the observation distance, and the variable u represents the spatial frequency. The graininess index can be calculated by the calculation formula shown in the following formula (2) based on the VTF. The coefficient τ is a coefficient for matching the obtained value with human senses. As apparent from the calculation method, it can be said that the granularity index is an index indicating whether or not a human feels that dots are conspicuous. It can be said that the smaller the value of the graininess index is, the more difficult it is to visually recognize dots in the print image quality, and the more excellent in that respect.

  VTF (u) = 5.05exp (-0.138πLu / 180) {1-exp (-0.1πLu / 180)} (1)

  Granularity index = τ∫FS (u) ・ VTF (u) du ・ ・ ・ （２）

  The initially prepared blue noise mask is configured to have the smallest granularity index, but if the threshold value is randomly changed in step S230, the granularity of the work mask is lower than that of the blue noise mask. To do. Therefore, a target range of the graininess index is set within a range that is acceptable in view of human visual characteristics, and it is determined whether there is no problem in view of this range. Of course, since the granularity index is a value determined for each gradation value, an upper limit value is prepared for each gradation value, and if the granularity index for each gradation value is less than or equal to this upper limit value, the granularity characteristic is within the target range. You may judge that it is in.

  If there is a problem with the graininess characteristic, that is, it is not within the target range and has not been improved compared to before the replacement of the threshold (step S260, "NO"), the process returns to step S230, and the threshold The above process is repeated from the replacement. As a result of repeating the processing of steps S230 to S260, when it is determined that the pair dot characteristics are improved and the graininess characteristics are satisfactory (steps S250 and S260: “YES”), the loop of steps S230 to S260 is temporarily performed. The process returns to step S220 to determine whether or not the pair dot generation characteristic is within the target range.

  If it cannot be determined that the pair dot generation characteristic is within the target range (“NO” in step S220), the processing from step S230 described above is repeated. In the process shown in FIG. 14, steps S220 to S260 are repeatedly executed while changing the threshold value until the condition is satisfied. Therefore, while the number of times that the processing from step S230 to S260 is executed (hereinafter referred to as the number of loops) is small, the upper limit value of the graininess index in step S260 is increased, and as the number of loops increases. Alternatively, processing such as bringing the upper limit value closer to the final target value may be performed. Thus, by changing the upper limit value according to the number of loops, it is possible to prevent the granularity index from falling into a local minimum value.

  In this way, the loop processing from step S230 to S260 is executed several times, and there is no problem in the graininess characteristic, and the right adjacent pair dot number RPD [1: 127] and the lower adjacent pair dot number UPD [1: 127] are the target. If it can be determined that it falls within the range M [1: 127] (step S220, “YES”), the paired pixel control mask is completed, and the current working mask is saved as a paired pixel control mask (step S270). END "is exited and the pair pixel control mask generation routine (FIG. 14) is terminated. In the above description, whether or not the pair dot generation characteristic is within the target range is performed in the range of 1 to 127/255 out of the entire range of gradation values in which dot generation is possible. The pixel control mask may be used in the gradation range in which the pair dot occurrence probability is to be controlled. For example, it may be performed only in a lower density range (tone range corresponding to dot generation probability k = 0 to 0.25, 0.2 to 0.5, etc.).

By the method described above, a pair pixel control mask can be obtained based on the blue noise mask. This dither mask is the dither mask used in the determination of dot formation in the first embodiment. Since this pair pixel control mask is based on a blue noise mask, analyzing the distribution of dots formed in the range where the gradation value of the image is low as a spatial frequency, it is almost in the low frequency region where human visual sensitivity is high. Have no ingredients. Therefore, it is possible to provide a dither mask that can realize high image quality. In addition, in the above-described pair pixel control mask, the probability of occurrence of a pair dot in which dots are formed in adjacent pixels is such that the dot formation probability k at that gradation value is about k ² × 0.8. Has been. As a result, even if a deviation occurs in the dot formation positions in the forward movement and the backward movement, the variation in the coverage is suppressed, and the change in the CIEL * a * b * color space caused by the deviation in the dot formation positions is suppressed. It is suppressed to the following range.
(A) When the image to be printed is yellow (RGB value, R = G = 255, B = 0-64, where RGB value is converted to 8 bits), the change in b * is less than or equal to 2 It is a range.
(B) When the image to be printed is magenta (RGB value, R = B = 255, G = 0-64, where RGB value is converted into 8 bits), the change in L * is 0.5. The following range or the range of a * is a range having a value of 0.5 or less.
(C) When the image to be printed is cyan (RGB value, G = B = 255, R = 0-64, where RGB value is converted into 8 bits), the change in L * is 0.5. The range is as follows.
Note that it is not necessary to satisfy all of these conditions, and it is sufficient that at least one of them is satisfied.
Thus, a dither mask capable of suppressing the occurrence of image unevenness is provided. If the dither mask made according to FIG. 14 does not satisfy the above condition, the target value of the coverage rate, specifically, the value of the coefficient to be multiplied by k ² (0.8 in the above example) is changed. Then, the process of FIG. 14 may be repeated until a desired characteristic is obtained.

In the present embodiment, the pair pixel control mask is generated using the blue noise mask as a starting point, but it can also be generated from a dither mask having other arbitrary characteristics. As described above, the time required for generation can be shortened by generating from a blue noise mask or a green noise mask that is excellent in dispersibility and whose original dispersibility is close to the desired characteristics. Further, when generating a dither mask from scratch, it is also possible to generate a pair pixel control mask by applying the following rule.
(1) The threshold values are sequentially arranged in the matrix from either the small side or the large side.
(2) When the next threshold value is arranged with respect to the threshold value already arranged at a certain position, the evaluation value such as the granularity index is used to associate the arrangement value of the next threshold value with the evaluation value in that case. . Then, candidates for the arrangement position of the next threshold value are identified in order of high evaluation.
(3) The above candidates are taken out in order from the highest evaluation side, and the number of paired dots in that case is counted. When a candidate having the required number of paired dots (for example, the number shown in FIG. 15) is found, the next threshold value is arranged at that position.
(4) Repeat (1) to (3) above until the threshold is exhausted.
Using such a rule, the arrangement of threshold values may be determined from the beginning, and a pair pixel control mask may be generated.

C. Modification of the first embodiment:
A modification of the first embodiment described above will be described. In the first embodiment, in the forward and backward movements, the dot formation positions are staggered in both the main scanning and sub-scanning directions, and the positions of adjacent pixels constituting the paired pixels are shown in FIG. As described above, the two pixels on the right side in the main scanning direction and the lower side in the sub-scanning direction are used. However, the adjacent pixels are not limited to these two, and the right and left adjacent pixels in the raster may be included on the lower side in the sub-scanning direction. good. When the position of the pixel of interest OJ is (0, 0), not only the pixels at the positions (1, 0) and (0, 1) as adjacent pixels, but also (-2, 1), (2, 1) The pixel at the position is also regarded as an adjacent pixel, and a total of four pairs of paired dots are counted. This is shown in FIG. Of course, this range can be further expanded to 8 pixels in FIG. 10C. If the range of the paired pixels is wide, it is possible to suppress the occurrence of density unevenness even if the manner of shifting the dot formation position is changed. It should be noted that it is desirable that the range of adjacent pixels be wide in the direction in which the deviation of the dot formation position in printing tends to occur. As shown in FIG. 10B, if adjacent pixels are set widely in the main scanning direction, it is effective to suppress the density due to the shift in the main scanning direction.

  Further, as shown in FIG. 3 (A), the dots formed during the backward movement of the dots formed during the forward movement can be alternated between the columns or the raster alternating as shown in FIG. 3 (B). Even in these cases, various settings can be made for the range of adjacent pixels.

  In the above-described embodiments and modifications, the resolution of the gradation value of the image is set to 8 bits and the threshold range is set to 0 to 255 in order to simplify the description, but the threshold value arranged in the dither mask is set to 0 to 0. If the number of bits representing the gradation value of the image is increased to 4095, for example to 10 bits, the number of dots arranged can be reduced with respect to the minimum gradation value 1, and the gradation value is 1 The number of dots that increase with each increase can be reduced. Therefore, the control of the occurrence probability of pair dots can be performed more finely. Of course, if the dither mask is further increased to 128 × 128, 256 × 512, or the like, the number of dots formed when the gradation value expressed by 10 bits is the value 1, the former is 4, and the latter It becomes about 32 pieces. The size of the dither mask, the number of bits representing the gradation value, the type of threshold value arranged in the dither mask, and the like are the purpose of the halftone processing to be performed (whether image quality priority or processing speed priority, or large format printing). Or the like) and the processing time.

  In the first embodiment, the paired pixel control mask is set assuming that the control of the paired dot is in the range of gradation values 0 to 127/255, in other words, the dot generation probability k is in the range of 0 <k <0.5. In this range, the upper limit value may be further limited to the lower gradation value side. For example, pair dot control may be performed only in a range of 0 <k <0.2. In general, as the actual dot size with respect to the pixel size increases, dot overlap due to misalignment is more likely to occur, so that the gradation region in which density variation is a problem moves to the low density side. Therefore, it is realistic to change the range to be adjusted according to the actual dot size with respect to the pixel size. Further, the lower limit value may be further limited to the higher gradation value side. In general, in the low gradation region near the gradation value 0, the dot formation positions are originally far apart, and the problem of density unevenness is not so obvious even if the dot formation positions are shifted. Therefore, the pair pixel control mask may be generated only in the range of 0.1 <k <0.4 or 0.2 <k <0.5. Further, the ratio of dots formed during forward movement and the ratio of dots formed during backward movement are differentiated from the beginning, and for each of them, different dot formation probabilities k1 and k2 are set to perform pair dot control. There is no problem.

D. Second embodiment:
D-1. Dither mask:
Next, a second embodiment of the present invention will be described. The hardware of the printer 20 as the second embodiment is the same as that of the first embodiment (see FIG. 1). The print control process (FIG. 4) in the printer 20 is the same except that the dither mask used in the halftone process is different. Also in the second embodiment, the halftone process is performed by a so-called dither method.

The difference between the dither mask used in the second embodiment and the dither mask used in the first embodiment is as follows.
(1) Common points of both:
In both cases, the size of the dither mask is 64 × 64.
Both are distributed dither masks that prioritize dispersibility.
When there is a shift in the dot formation position in forward and backward movement, the change in the CIEL * a * b * color space satisfies the following conditions (A) to (C).
(A) When the image to be printed is yellow (RGB value, R = G = 255, B = 0-64, where RGB value is converted to 8 bits), the change in b * is less than or equal to 2 It is a range.
(B) When the image to be printed is magenta (RGB value, R = B = 255, G = 0-64, where RGB value is converted into 8 bits), the change in L * is 0.5. The following range or the range of a * is a range having a value of 0.5 or less.
(C) When the image to be printed is cyan (RGB value, G = B = 255, R = 0-64, where RGB value is converted into 8 bits), the change in L * is 0.5. The range is as follows.
(2) Differences between the two:
The first dither mask 61 (not shown) having excellent dot dispersibility is once generated, and then the first dither mask 61 is shifted to generate the second dither mask 62. The threshold value of the 64 × 64 dither mask in the first embodiment was created considering only the dispersibility of the distribution of dots formed by the forward and backward movements of the print head 90, whereas In the second embodiment, the first dither mask 61 includes a first pixel group to which dots formed when the print head 90 moves forward and a second pixel group to which dots formed when the print head 90 moves backward. Is made in consideration of the dispersibility of the dots.

  Also in the second embodiment, since the second dither mask 62 satisfies the above conditions (A) to (C), the positions of dots formed when the print head 90 moves and the backward movement Even if there is a shift in the position of the dots that are sometimes formed, the coverage does not change greatly, and the change in the CIEL * a * b * color space of the image is sufficiently suppressed. The effect is almost the same as the example. In particular, when printing is performed at a reference position in which the position of dots formed during forward movement and the position of dots formed during backward movement are sufficiently adjusted, 2/720 inch to 5/720 inch from the reference position. Even when compared with a case where printing is performed with a deviation of an arbitrary distance within the range of, the range of change in the L * a * b * color space is sufficiently small and recognized as a change in brightness or color. It will not be done.

D-2. Generation method of the first dither mask 61 and the second dither mask 62:
A method of generating the first dither mask 61 and the second dither mask 62 described above will be described. The first dither mask 61 and the second dither mask 62 have storage elements corresponding to their sizes (number of threshold values). In this embodiment, the size of the first dither mask 61 and the second dither mask 62 is the same. The storage element is an element that stores a threshold value constituting the first dither mask 61 and the second dither mask 62. The first dither mask 61 and the second dither mask 62 are generated by storing threshold values one by one in all of these storage elements. The first dither mask 61 is a dither mask generated in the middle of the generation process of the second dither mask 62. The generation method described below is a process of generating the first dither mask 61 and the second dither mask 62 by a CPU such as a main frame. It should be noted that some or all of the steps described below may be performed manually by the user. The first dither mask 61 is used for generating the second dither mask 62 and does not need to be stored in the EEPROM 60 of the printer 20. The first dither mask 61 is normally stored in a memory in an apparatus (computer) that generates the second dither mask 62.

  The size of the first dither mask 61 and the second dither mask 62 can be, for example, 256 pixels × 256 pixels, 512 pixels × 512 pixels, etc. In the following description, the first dither mask 61 In order to simplify the description, the size of the second dither mask 62 will be described as a size applied to image data of 5 pixels in both the vertical size and the horizontal size, that is, a total of 25 pixels (= 5 × 5). .

  FIG. 17 is a process diagram showing the procedure of the method for generating the first dither mask 61 and the second dither mask 62. In the generation of the first dither mask 61 and the second dither mask 62, as shown in the figure, first, threshold values corresponding to the sizes of the first dither mask 61 and the second dither mask 62 are prepared (step S310). ). In the present embodiment, since the first dither mask 61 and the second dither mask 62 have 25 storage elements, the same number of threshold values 0 to 24 are prepared.

  Once the threshold value is prepared, a target threshold value selection process is performed (step S320). The target threshold value selection process is a process of selecting one threshold value from among the prepared threshold values 0 to 24 as a target threshold value from among threshold values not yet stored in the storage element. In this embodiment, the threshold value of interest is selected in order from the smallest threshold value of the prepared threshold values. As shown in FIG. 19, in the storage elements constituting the first dither mask 61 and the second dither mask 62, when threshold values 0 to 3 are already stored in the storage elements by the process described later, Next, the threshold value of interest selected in step S320 is the value 4.

  If the target threshold value is selected, then a dither mask evaluation process is performed (step S330). The dither mask evaluation process is a storage in which a threshold is already stored when a target threshold is stored for one of the storage elements for which a prepared threshold is not yet stored (hereinafter also referred to as a blank storage element). This is a process of calculating, for each blank storage element, an evaluation value E indicating the degree of dot dispersion for the dot formation pattern represented by the arrangement of elements (hereinafter also referred to as “determination storage elements”).

  In this embodiment, the smaller the value, the better the dot dispersibility, and it can be said that the evaluation value E is superior from the viewpoint of the granularity of the printed image. In the halftone process, since the dots are turned on at pixels where the threshold values of the first dither mask 61 and the second dither mask 62 are smaller than the input gradation value, the gradation values of all the pixels are the same. In the case of inputting a solid image, if the gradation value is gradually increased, a dot formation pattern corresponding to the arrangement of threshold values in the first dither mask 61 and the second dither mask 62 appears. Become. In this embodiment, the dot shape based on such dot generation characteristics is called a dot formation pattern.

  Details of the dither mask evaluation processing will be described with reference to FIG. In the dither mask evaluation process, as shown in FIG. 18, first, a grouping process is performed (step S331). The grouping process refers to a plurality of storage elements constituting the first dither mask 61 and the second dither mask 62 at dot formation positions where threshold values stored in the plurality of storage elements are applied in the halftone process. In this process, the dots are formed by focusing on whether the dot formation is performed in the forward movement or the backward movement. That is, this is a process for setting a group of storage elements based on the arrangement of the forward and backward dots described above. For example, when the column-alternating dot arrangement is adopted, the pattern is divided into groups having the same pattern as the forward pixel group and the backward pixel group shown in FIG.

  Thus, when the grouping process is performed, the dot of the decision storage element is turned on (step S332). In FIG. 19, the state in which the dots of the decision storage elements storing the threshold values 0 to 3 are turned on is indicated by single hatching. If the dot of the determined storage element is turned ON, candidate storage element selection processing is then performed (step S333). The candidate storage element selection process is a process of selecting a candidate storage element that is a candidate for a storage element in which a target threshold value is to be stored. Since it is possible to store the target threshold value in each blank storage element, one of the blank storage elements is selected as a candidate storage element here. Once the candidate storage element selection process is performed, the dot of the candidate storage element is then turned ON (step S334). In FIG. 19, a state where one of the blank storage elements is selected as a candidate storage element and the dot of the candidate storage element is turned ON is shown by cross hatching.

  If the dot of the candidate storage element is turned on, next, group selection processing is performed (step S335). The group selection processing includes p groups G1 to Gp (p is an integer of 2 or more, here p = 2) set in step S331, a first dither mask 61, and a second dither mask 62. This is a process of selecting one group Gq (q is an integer not less than 1 and not more than p + 1) from the group Gp + 1 that is a group including all storage elements.

  When the group Gq is selected, the evaluation value Eq indicating the degree of dot dispersion, that is, how much the dots are uniformly distributed is formed based on the dot formation pattern corresponding to the storage elements belonging to the group Gq. An evaluation value indicating whether or not is calculated (step S336). It is known that a dither mask having a blue noise characteristic shown in FIG. 20 may be generated in order to form dots uniformly dispersed. The blue noise characteristic is a noise characteristic in which the dot distribution has a peak on the high frequency side in the spatial frequency region. As for human visual characteristics, when the observation distance is 30 cm, the sensitivity at a high frequency of 10 cycles / mm or more is almost zero, so that a dither mask with good graininess can be generated if it has a peak in a region beyond that. . In this embodiment, in order to generate a dither mask having such characteristics, the graininess index is used as an evaluation value indicating the degree of dot dispersibility, as in the first embodiment (see FIG. 16 and FIG. 16). Formulas (1) and (2)).

When the evaluation value Eq is calculated, the steps S335 and S336 are repeated until the evaluation values Eq are calculated for all the groups G1 to Gp + 1 (here, G1 to G3) (step S337). Thus, when the evaluation values Eq (E1 to E3) are calculated for all the groups G1 to G3 (step S337: YES), the evaluation value E is calculated by the following equation (3) based on the calculated evaluation values E1 to E3. (Step S338). In Expression (3), a to c are weighting coefficients. These weighting factors are experimentally determined as constant values so that good print image quality can be obtained. That is, the evaluation value E is the dot formation pattern represented by the entire determination storage elements of the first dither mask 61 and the second dither mask 62, and the respective dot formation patterns represented by the determination storage elements corresponding to the forward movement. The evaluation value obtained by comprehensively evaluating the degree of dot dispersion for each dot formation pattern represented by the decision storage element corresponding to the backward movement with a predetermined weight.
E = a × E1 + b × E2 + c × E3 (3)

  When the evaluation value E is calculated, the steps S333 to S338 are repeated until the evaluation value E is calculated for all candidate storage elements (blank storage elements) (step S339). When the evaluation value E is calculated for all candidate storage elements in this way (step S339: YES), the dither mask evaluation process is completed.

  When the above-described dither mask evaluation process (FIG. 18) is completed, a process for determining a storage element is executed (FIG. 17, step S340). Here, the storage element determination process is a process of determining a blank storage element for storing the target threshold value based on the evaluation value E calculated in the dither mask evaluation process. Specifically, among the evaluation values E calculated for each candidate storage element (blank storage element), the candidate storage element corresponding to the evaluation value E having the smallest value is set as the candidate storage element for storing the threshold value of interest. The threshold value of interest is stored in the storage element.

  When the storage element determination process is performed, the processes of steps S320 to S340 are repeated until the threshold value prepared in step S310 is stored in all the storage elements constituting the first dither mask 61 and the second dither mask 62. (Step S350). Thus, when threshold values are stored in all storage elements (step S350: YES), first, the first dither mask 61 is completed (step S360).

  The first dither mask 61 generated using the evaluation value E described above is generated so as to generate a dot formation pattern in which dots are difficult to be visually recognized by weighting corresponding to the sensitivity characteristic VTF with respect to the visual spatial frequency that humans have. . Therefore, as a result, the dot formation pattern generated by the first dither mask 61 has a noise characteristic in which the dot distribution has a peak on the high frequency side in the spatial frequency region. In other words, the dot formation pattern generated by the first dither mask 61 has a blue noise characteristic. Therefore, the print image printed by the halftone process using the first dither mask 61 can obtain a smooth print quality in which the dots are well dispersed and the dots are hardly noticeable.

  Further, if the above-described evaluation value E is used, dot formation in which dots are dispersedly arranged for all of the forward movement dots, backward movement dots, and the entire print image in which the forward movement dots and the backward movement dots are combined. A first dither mask 61 having a pattern can be generated. If halftone processing is performed using the first dither mask 61, the dispersibility of the forward and backward dots remains secured even if a positional deviation occurs between the forward and backward dots. Therefore, the dispersibility of the dots of the entire image is ensured, and the deterioration of the graininess of the print image quality can be suppressed.

  When the first dither mask 61 is completed in this way, threshold value shifting processing is performed (step S370). The threshold value shifting process is a process for shifting the threshold values corresponding to some of the groups set in step S331 with respect to the threshold arrangement of the first dither mask 61 in a predetermined direction by a predetermined number of pixels. . By performing this threshold value shifting process, the second dither mask 62 is completed (step S380). Thus, the dither mask generation process ends.

  Details of the threshold value shifting process will be described below. The threshold arrangement of the first dither mask 61 is shown in FIG. The first dither mask 61 is a dither mask in which the dispersibility of dots is optimized by the above-described method, and is also called a dispersibility optimum dither mask. FIG. 21 shows a threshold arrangement of a dither mask having a size of 7 × 7 pixels. In order to distinguish the threshold values of each pixel, numbers such as “0, 1” are given to the respective threshold values in the drawing. For example, the threshold value “0, 1” indicates the 0th threshold value in the main scanning direction (right direction in the figure) and the 1st threshold value in the sub-scanning direction (downward direction in the figure). Each threshold value constituting the first dither mask 61 includes a threshold value applied to pixels in which dots are formed by forward movement (hereinafter also referred to as forward movement threshold values), and a threshold value applied to pixels in which dots are formed by forward movement. (Hereinafter also referred to as a backward movement threshold). In FIG. 21, the backward movement threshold value is displayed with hatching. The threshold arrangement of the first dither mask 61 shown in the figure is the case of dot arrangement of alternating columns.

  FIG. 22 shows a specific example of the threshold arrangement of the second dither mask 62 generated by the threshold shift process with respect to the threshold arrangement of the first dither mask 61. As shown in the drawing, the threshold arrangement of the second dither mask 62 is such that only the backward threshold value of the forward threshold value and the backward threshold value of the first dither mask 61 is set in the sub-scanning direction (upward in the figure). The threshold arrangement is shifted by 3 pixels. For this reason, the second dither mask 62 is also referred to as a shifted dither mask. Since the second dither mask 62 is repeatedly applied in the main scanning direction and the sub-scanning direction, for example, when the backward movement threshold value is shifted upward by 3 pixels, “1,0”, “1,1”, for example. , “1, 2” is placed below “1, 7”.

  Another example of the threshold arrangement of the second dither mask 62 is shown in FIG. In this example, as shown in the figure, the threshold arrangement of the second dither mask 62 is such that only the backward threshold of the first dither mask 61 is shifted by 4 pixels in the main scanning direction (left direction in the figure). The threshold arrangement is shifted by 3 pixels in the sub-scanning direction (upward in the figure).

  Such threshold value shift processing is applied to raster alternate dot arrangement (FIG. 3B) and task dot arrangement (FIG. 3C) in the same manner as the above-described column alternate dot arrangement. be able to. FIG. 24 shows an example of threshold arrangement of the second dither mask 62 when the raster alternate dot arrangement is adopted. In the illustrated example, the backward movement threshold value of the first dither mask 61 is shifted by 3 pixels in the main scanning direction (left direction in the figure). FIG. 25 shows an example of threshold arrangement of the second dither mask 62 in the case where the dot arrangement of tasks is adopted. In the illustrated example, the backward movement threshold value of the first dither mask 61 is shifted by 4 pixels in the main scanning direction (left direction in the figure).

  As described above, the direction in which the threshold value is shifted in the threshold value shifting process may be the main scanning direction or the sub-scanning direction. Of course, the direction may be the main scanning direction and the sub-scanning direction. Further, the threshold value to be shifted is not limited to the backward movement threshold value but may be the forward movement threshold value. By performing the threshold value shifting process in this way, the characteristics of the second dither mask 62 are changed in a direction in which the dot dispersibility is deteriorated as compared with the first dither mask 61. This means that the pair dot occurrence rate K is larger than that of the first dither mask 61. That is, the same characteristics as in the first embodiment can be obtained.

Further, the shift amount of the threshold value in each direction may be set as appropriate, but is preferably set as follows. Examining the characteristics of the blue noise mask, when the relative positions of two pixels in one dither mask are close, there is a strong correlation between ON / OFF of dots in the two pixels. That is, when the dot of one pixel of the two pixels is turned on, the probability that the dot of the other pixel is turned on is extremely low. On the other hand, when the two pixels are sufficiently separated from each other, there is almost no correlation between dot ON / OFF in the two pixels. When the relative position between the two pixels increases, the probability that both dots of the two pixels are turned on, that is, the pair dot generation rate K is k ² (k is the dot generation rate) as in the case of white noise. ) (See FIG. 11).

The relationship between the relative position between the two pixels and the paired dot occurrence rate K is such that the printing gradation in which density change is a problem due to the positional deviation, that is, from the low gradation to the middle gradation area (for example, 25 to 127/255). ) Was confirmed to converge to k ² when the relative position between the two pixels is about 5 pixels or more. This means that if the shift amount of the threshold is 5 pixels or more, pairs dot generation ratio K, which means that almost converges to k ^2. On the other hand, the threshold shift amount is preferably as small as possible in order to ensure good dot dispersibility of the second dither mask 62. This is because the granularity of the dot optimized as the first dither mask 61 deteriorates as the shift amount of the threshold increases.

Therefore, the shift amount of the threshold value, relative value pair dot generation ratio K is substantially converged to k ² (5 pixels in the example above), and sets the added amount of positional deviation is assumed values as the upper limit It is desirable. For example, when the actual positional deviation amount is assumed to be ± 2 pixels, it is desirable that the threshold shift amount has an upper limit value of 7 pixels. In this case, for example, the threshold shift amount may be 5 pixels. In this way, when the positional deviation of ± 2 pixels occurs, the substantial threshold deviation amount falls within the range of 3 to 7 pixels, and the dot dot generation rate K is suppressed while suppressing the deterioration of dot granularity. Can be brought close to k ² well. However, when the assumed positional deviation amount is larger, the threshold deviation amount needs to be further increased. For example, when the actual positional shift amount is −5 pixels, even if the threshold shift amount is 5 pixels, the substantial threshold shift amount is 0 pixels.

  The threshold shift amount is preferably set in consideration of the peculiarity of the positional shift of the printer 20. Specifically, when the positional deviation of the printer 20 is likely to occur in a specific direction, the shift amount in a direction different from the specific direction is shifted so as to be larger than the shift amount in the specific direction. It is desirable to set the amount. In the present embodiment, since the printer 20 is a serial printer, the positional deviation between the forward movement dot and the backward movement dot is likely to occur in the main scanning direction. Therefore, it is desirable to make the shift amount in the sub-scanning direction larger than that in the main scanning direction. In this way, the range of fluctuation of the total amount of the threshold shift amount and the actual displacement amount, that is, the substantial threshold displacement range is closer to the case where no displacement occurs. , Dot dispersion characteristics are stable. Also, the threshold shift amount can be reduced.

  For example, consider a case where a positional deviation of ± 2 pixels is assumed in the main scanning direction. As a first case, if the threshold shift amount is 6 pixels in the main scanning direction and 0 pixel in the sub-scanning direction, the fluctuation range of the total amount of the threshold shift amount and the actual positional deviation amount is as follows. The range is 4 to 8 pixels in the scanning direction and 0 pixel in the sub-scanning direction. On the other hand, as a second case, if the threshold shift amount is 0 pixel in the main scanning direction and 4 pixels in the sub-scanning direction, the fluctuation range of the total amount of the threshold shift amount and the actual positional deviation amount is The range is -2 to 2 pixels in the main scanning direction and 4 pixels in the sub-scanning direction. Comparing Case 1 and Case 2, Case 2 has a smaller total shift amount and suppresses deterioration of graininess. Moreover, in case 1 and case 2, a substantial threshold shift amount of 4 pixels or more can be secured. From the above, when the position shift is likely to occur in the main scanning direction, it is desirable to make the shift amount in the sub-scanning direction larger than the shift amount in the main scanning direction.

D-3. effect:
The printer 20 having such a configuration uses two dither masks depending on the moving speed of the carriage 80. If it is determined that the displacement of the dot formation position is unlikely to occur due to the slow movement speed of the carriage 80, the dot generated by the first dither mask 61 set to have a noise characteristic having a peak on the high frequency side. Print according to the data. As described above, by using the first dither mask 61 that emphasizes the graininess rather than the countermeasure against the positional deviation of the dots, higher quality printing can be performed.

  On the other hand, when it is determined that the displacement of the dot formation position is likely to occur due to the high movement speed of the carriage 80, the forward pixel group and the backward pixel group are compared with the threshold arrangement of the first dither mask 61. Printing is performed according to the dot data generated by the second dither mask 62 in which the threshold values corresponding to some of the groups are shifted by a predetermined pixel in a predetermined direction. For this reason, even when dot misregistration occurs between pixel groups, the misalignment direction is set so that the change in the CIEL * a * b * color space of the image is within a predetermined range, as in the first embodiment. It is possible to suppress fluctuations in the dot coverage according to the amount. As a result, it is possible to suppress the occurrence of density unevenness in the print image in the common area and the deterioration of the print image quality. Moreover, the threshold arrangement of the first dither mask 61, which is the basis of the second dither mask 62, has a noise characteristic having a peak on the high frequency side in the spatial frequency domain, and the printed image is at least common. In the region, a characteristic close to a so-called blue noise characteristic is exhibited, and a predetermined degree of granularity can be ensured.

An example of a specific effect of the printer 20 is shown in FIG. FIG. 26 shows a simulation result. Specifically, cases in which various amounts of Bi-D deviation occur in the main scanning direction and the sub-scanning direction are set for a dot formation pattern using a dither mask that exhibits good blue noise characteristics. The coverage is obtained and plotted for these cases. The horizontal axis in FIG. 26 is the total value of the positional deviation amounts in the main scanning direction and the sub-scanning direction (hereinafter also referred to as the total deviation amount). The prerequisites for the simulation are as follows.
(1) Tone value: 80/255
(2) Interlace processing: Arranged dot layout (3) Dot size: The dot diameter is twice the size between pixels (4) Amount of displacement:

  Regarding the positional deviation amount, as the deviation amount in the main scanning direction, 0 pixel and 0.5 to 5.5 pixels were respectively adopted at 1-pixel intervals, and seven kinds of deviation amounts were set. In addition, as the amount of deviation in the sub-scanning direction, between 0 and 4 pixels were adopted at 0.5 pixel intervals, and nine different amounts of deviation were set. FIG. 9 plots the coverage of cases where the total deviation amount is less than 7 pixels (51 patterns) among Bi-D deviation cases of all combinations (63 patterns) of these deviation amounts.

As shown in FIG. 26, when the case where the total deviation amount is 0 pixel and the case where it is 2 pixels are compared, a difference of 10% or more is generated in the coverage. On the other hand, when the total deviation amount becomes larger than 2 pixels, it can be seen that the variation amount of the coverage ratio decreases as the total deviation amount increases. Comparing the case where the total deviation amount is 2 pixels and the case of 3 pixels, the variation in the coverage is about 2%. When the total deviation exceeds 5 pixels, the change in the coverage is very stable, and even if the total deviation is increased, the coverage is hardly changed. This corresponds to the fact that the pair dot generation rate K almost converges to k ² when the shift amount of the threshold when generating the second dither mask 62 is 5 pixels or more.

  Here, if the shift amount of the threshold when generating the second dither mask 62 is 5 pixels, if the actual displacement amount is ± 2 pixels, the substantial displacement that appears in the printed image. The range of the quantity is in the range of 3 to 7 pixels. In this range, the fluctuation range of the coverage is less than 2%. On the other hand, when the shift amount of the threshold when generating the second dither mask 62 is 0 pixel, that is, when the conventional first dither mask 61 having the blue noise characteristic is used, the actual positional shift amount is In the case of ± 2 pixels, the range of the substantial misregistration amount appearing in the print image is a range of −2 pixels to 2 pixels. In this range, the fluctuation range of the coverage is as high as 14% or more.

  As is clear from the above description, the second dither mask 62 greatly suppresses the variation in the coverage and suppresses the change in the CIEL * a * b * color space of the printed image within a predetermined range. can do. The local variation of the coverage means the occurrence of density unevenness in the printed image. Therefore, if the variation in the coverage of the printed image is suppressed using the second dither mask 62, the occurrence of density unevenness can be significantly suppressed.

  The threshold arrangement of the second dither mask 62 in which the threshold is shifted with respect to the threshold arrangement of the first dither mask 61 has the same influence on the print image quality as the positional deviation that occurs when the first dither mask 61 is used. Therefore, the graininess is deteriorated as compared with the first dither mask 61. However, the threshold arrangement of the first dither mask 61 that is the basis of the second dither mask 62 is set to have a noise characteristic having a peak on the high frequency side in each of the forward pixel group and the backward pixel group. Therefore, the print image printed using the second dither mask 62 can suppress such deterioration. That is, the second dither mask 62 can significantly suppress the occurrence of density unevenness while suppressing the deterioration of graininess.

  As described above, the second dither mask 62 described above is generated by shifting threshold values corresponding to some pixel groups among the threshold values of the first dither mask 61 having good blue noise characteristics. Therefore, the second dither mask 62 has a predetermined property. That is, when a plurality of dither masks are generated by shifting the thresholds corresponding to some pixel groups in various directions by various amounts in the threshold arrangement of the second dither mask 62, among these dither masks, A dither mask shifted by a certain amount in a certain direction has better blue noise characteristics than the second dither mask 62, and the printed image has the property of improving the graininess. In other words, the second dither mask 62 has a spatial frequency characteristic closer to that of the blue noise mask than to the white noise mask as much as possible while maintaining the paired dot generation rate K that is close to that of the white noise mask. It can also be called a dither mask.

E. Modification of the second embodiment:
A modification of the above embodiment will be described.
E-1. Modification 1:
In the second embodiment, in step S331, the pixel group of the print image is divided into a pixel group consisting of forward movement dots and a pixel group consisting of backward movement dots, and the threshold value of the first dither mask 61 is determined. A configuration is shown in which the second dither mask 62 is generated by shifting the threshold corresponding to one of the pixel groups with respect to the arrangement by a predetermined pixel in a predetermined direction. However, the pixel group to be divided is not limited to the above example, and may be any pixel group with different printing conditions.

  For example, when the print head 90 prints an image of a predetermined area by a plurality of main scans, the pixel group of the print image is a pixel group in which dots are formed by different main scans of the plurality of main scans. It may be divided into groups. For example, the region of interest shown in FIG. 3 is configured by repeating regions in which an image is completed by passes 1 to 4. Therefore, the pixel group of the print image may be divided into four pixel groups 1 to 4 corresponding to each pass. In this case, in the threshold value shifting process, the threshold value corresponding to at least one of the other pixel groups 2 to 4 may be shifted on the basis of one pixel group, for example, the pixel group 1. For example, the threshold value corresponding to the pixel group 2 is 4 pixels in the main scanning direction, the threshold value corresponding to the pixel group 3 is 4 pixels in the sub scanning direction, and the threshold value corresponding to the pixel group 4 is 2 pixels in the main scanning direction. May be shifted by two pixels. The direction in which the threshold value is shifted and the shift amount may be set as appropriate in consideration of the occurrence characteristics of the positional deviation between the paths.

E-2. Modification 2:
In the second embodiment, the evaluation value E is calculated in all print gradation ranges and the threshold arrangement of the first dither mask 61 is determined. The threshold arrangement may be determined based on E. For example, the threshold arrangement of the first dither mask 61 may be set by another method in a high gradation printing region where dot dispersibility is not a significant problem.

E-3. Modification 3:
In the second embodiment, the first dither mask 61 and the second dither mask 62 are stored in advance in the EEPROM 60. However, the present invention is not limited to such a configuration. For example, the EEPROM 60 may store one of the first dither mask 61 and the second dither mask 62 in advance. In this case, the CPU 40 may function as a dither mask generation unit and generate the other from one of the first dither mask 61 and the second dither mask 62. Of course, in the halftone process, it is not essential to select a dither mask according to the type of print medium, and the printer 20 stores only the second dither mask 62, regardless of the type of print medium. The second dither mask 62 may be fixedly used, or a dither mask different from the first dither mask 61 and the second dither mask 62 may be switched and used.

E-4. Modification 4:
In halftone processing, when a dither mask is repeatedly applied in the main scanning direction, the application position of the dither mask may be shifted by a predetermined amount in the sub-scanning direction. The threshold value shifting process in such a case will be described. FIG. 20 shows an example of threshold arrangement of the second dither mask 62 that has been subjected to the threshold shifting process. In this example, as shown in the figure, when the second dither mask 62 is repeatedly applied in the main scanning direction a plurality of times, the application position of the dither mask is shifted by one pixel in the sub scanning direction. .

  In this example, the backward threshold is shifted by 4 pixels in the main scanning direction (left direction in the figure) with respect to the first dither mask 61. In addition to the shift, the threshold values “1, x” and “3, x” (x is an integer of 0 to 7) among the backward movement threshold values are set to one pixel in the sub-scanning direction (upward in the figure). It ’s a minute away. This is because the first dither mask 61 is optimized on the assumption that the first dither mask 61 is shifted by a predetermined amount in the sub-scanning direction. Specifically, in the first dither mask 61, for example, focusing on the backward movement threshold, in the first dither mask 61, the backward movement threshold next to “7, 0” is “1, 1”. Optimized for. In the threshold arrangement shown in FIG. 27, the threshold values “1, x” and “3, x” of the backward movement threshold values are set to 1 in the sub-scanning direction in order to maintain the relative relationship of the backward movement threshold values. The pixel is shifted. In this way, since the preconditions of the first dither mask 61 are maintained, it is possible to suppress the deterioration of graininess, which is desirable.

E-5. Modification 5:
In the second embodiment, the granularity index is used as the evaluation value Eq indicating the degree of dot dispersion. However, the evaluation value Eq may be any value that can evaluate the degree of dispersion of the dot arrangement. For example, RMS granularity may be used as the evaluation value Eq. Since RMS granularity is a known technique (for example, Japanese Patent Application Laid-Open No. 2007-174272), detailed description is omitted, but the low-pass filter processing is performed on the dot density value using a low-pass filter, and the low-pass filter is used. The standard deviation of the processed density value is calculated. Alternatively, the evaluation value Eq may be a dot density after appropriate low-pass filter processing, such as a weight value filter that is inversely proportional to the distance from the pixel position.

E-6. Modification 6:
In the above-described embodiment, an example in which the printing apparatus of the present invention is realized as a serial printer has been described. However, the present invention performs dot formation in a plurality of pixel groups with different printing conditions, The present invention can be widely applied to printing apparatuses that output a print image by performing at least a part of dot formation by a plurality of pixel groups in a common area. The different printing conditions can be differences in the timing, position, print head, and nozzle for forming dots.

  For example, the printing apparatus of the present invention can also be realized as an inkjet line printer. Specifically, for example, a line printer including a plurality of print heads arranged over the print range and forming dots on the print medium may be used. In such a line printer, adjacent print heads among a plurality of print heads may be arranged so that a part thereof overlaps in the direction of the print head arrangement. In this overlapping area, the dots formed by the two print heads are combined to complete a printed image. For such a line printer, in step S331 described above, each of the groups of pixels in which dots are formed by different print heads among the plurality of print heads may be divided.

  Alternatively, it may be a line printer in which nozzles are arranged over the printing range and a plurality of nozzle rows that eject ink are provided in a crossing direction that intersects the direction of the nozzle row arrangement, that is, in the paper feed direction. Such a line printer can improve the printing resolution by providing a plurality of nozzle rows. That is, the print image is completed by combining dots formed by the two nozzle rows. If the nozzle positions in the nozzle arrangement direction are the same among the plurality of nozzle arrays, the printing resolution in the paper feed direction can be improved. Further, if the nozzle positions in the nozzle arrangement direction are different among the plurality of nozzle arrays, the printing resolution in the nozzle arrangement direction can be improved. For such a line printer, in step S331, the line printer may be divided into groups of pixels in which dots are formed by different nozzle rows of the plurality of nozzle rows. In such a case, the plurality of nozzle rows may be provided in one print head, or different print heads may be provided individually.

  When the nozzles constituting one nozzle row are arranged in a zigzag shape in the nozzle row arrangement direction, the positions of the two nozzle groups forming the zigzag shape, that is, in the paper feed direction, are different. You may divide into each group of the pixel in which a dot is formed by one nozzle group.

  In these line printers, misalignment tends to occur in the paper feed direction. Therefore, in the generation of the second dither mask 62 used in the line printer, it is preferable to relatively increase the shift amount in the direction different from the paper feed direction, that is, in the direction intersecting the paper feed direction, in the threshold value shift process. The effect of the present invention is particularly exerted when applied to a printer having a printing speed of 6 inches / second or more as a second setting or 0.5 sheets / second or more of A4 paper.

E-7. Modification 7:
In the above-described embodiment, the printer 20 is configured to perform printing alone, but in a printing system (broadly defined printing apparatus) in which a printer and a computer (a computer as a terminal, a computer as a print server, etc.) are connected. When the printing process is performed, the computer may perform a part of the printing process illustrated in FIG. For example, dot data that has been halftoned by a RIP (Raster Image Processor) may be printed by the printer 20. A dedicated hardware RIP may be used. For example, the hardware RIP executes halftone processing on data transferred from a computer, and transfers dot data generated thereby to a printer.

E-8. Modification 8:
The moving speed of the carriage 80 may be changed to any numerical value. However, 200 cps or more is preferable. This is because when it exceeds 200 cps, especially when it is 400 cps or more, the dot formation position is liable to shift, and the effects of the present invention are particularly exhibited.

  As mentioned above, although embodiment of this invention was described, elements other than the element corresponding to the element described in the independent claim among the components in embodiment mentioned above are additional elements, are abbreviate | omitted suitably, or Combinations are possible. Further, the present invention is not limited to such an embodiment, and it is needless to say that the present invention can be implemented in various modes within a range not departing from the gist of the present invention. For example, the present invention can be realized not only as a printing apparatus but also as a dither mask generation method, a printing apparatus manufacturing method, and the like. In addition, the present invention can be realized as a printing apparatus or the like that performs monochrome printing or sepia printing instead of the configuration that performs multicolor printing. In this case, a print head having a nozzle row that ejects single-color ink droplets may be used.

  Furthermore, the present invention can be implemented as a printer that can form ink droplets of two or more sizes such as large, small, large, medium, and small as ink droplets. In order to control the size of the ink droplet, for example, in a printer having a print head of a type in which the volume of the ink chamber is controlled by an actuator such as a piezo element, the actuator is controlled and the ink interface (meniscus) of the nozzle tip is controlled. What is necessary is just to control a state. Alternatively, in a printer equipped with a print head that heats ink with a heater provided in the nozzle and discharges ink droplets from the nozzle by the generated bubbles, the number of heaters driven and power are controlled to grow bubbles. Should be controlled.

  As described above, the present invention can also be applied to a printing apparatus and a printing method that perform dot concentration type dot formation. In a dot-concentrated printing device, the number of dots formed in a concentrated manner increases as the gradation value of the image to be printed increases. Therefore, when printing is divided into a plurality of pixel groups, each pixel group When the printing position is shifted, the number of overlapping dots decreases, and the coverage rate generally increases. For this reason, the printed image changes in brightness and hue as compared with the case where there is no shift in the printing position between pixel groups. Therefore, similarly to the dot dispersion type dither mask described above, the threshold arrangement in the dot concentration type dither mask is determined so as to suppress the fluctuation of the dot overlap ratio when the dot formation position is shifted. It ’s fine. The degree of suppression (color difference change range) is the same as in the case of dot dispersion type dot formation.

  When a dot-concentrated dither mask is used, the probability of paired dots is set higher in a range where the number of dots to be formed is small (high brightness range) than in the case where dots are randomly formed. Therefore, the arrangement of the dither mask thresholds is corrected by the same processing as the processing shown in FIG. 14 so that the probability of the paired dots approaches the probability of forming the paired dots when the dots are randomly formed. It is.

  The printing apparatus and the like of each of the embodiments described above has L *, a *, and b * in the CIEL * a * b * color space of the printed image because specific characteristics are given to the dither mask. The change is suppressed within a predetermined range. Therefore, it is possible to determine whether the present invention is implemented without necessarily analyzing the characteristics of the dither mask. In other words, when a dither mask with high dispersibility is used, as shown by the symbol N in FIGS. 7 to 9, if a deviation occurs in dot formation positions in forward and backward movements, CIEL * a * b * In the color space, L *, a *, and b * of the image change beyond a predetermined range. At least one change of L *, a *, and b * indicates that the printing state changes from the setting of the inter-dot distance 0 of the dots of the first and second pixel groups to an inter-dot distance of 2/720 inch to 5 / If the L * is within 2.0, a * is within 0.5, and b * is within 0.5 when the setting is shifted to an arbitrary value within the range of 720 inches, the present invention is implemented. Can be determined. Such shift of the dot formation position can be easily realized by adjusting the shift of the dot formation position between forward movement and backward movement in bidirectional printing, for example. Therefore, it is possible to easily determine whether or not the present invention is implemented by measuring changes in brightness and color of two images printed by changing the dot formation position in bidirectional printing.

20 ... Printer 30 ... Control unit 40 ... CPU
42 ... Halftone processing unit 43 ... Printing unit 51 ... ROM
52 ... RAM
60 ... EEPROM
DESCRIPTION OF SYMBOLS 61 ... 1st dither mask 62 ... 2nd dither mask 70 ... Carriage motor 71 ... Drive belt 72 ... Pulley 73 ... Sliding shaft 74 ... Paper feed motor 75 ... Platen 80 ... Carriage 82-87 ... Ink cartridge 90 ... Printing Head 98 ... Memory card slot 99 ... Operation panel P ... Printing paper MC ... Memory card

Claims (5)

A printing apparatus that performs printing by forming dots on a print medium,
A print head for performing color printing with multi-color dots at a resolution N (N is a natural number of 180 or more) dpi (dot per inch);
When forming dots on the print medium by the print head according to dot data representing the presence or absence of dot formation for each pixel, the dots are formed by dividing the dots into at least first and second pixel groups, A printing unit that prints an image by performing at least part of dot formation by the first and second pixel groups in a common area;
The printing unit performs printing according to the first setting when the speed of the print head with respect to the print medium is lower than a reference speed, and performs printing according to the second setting when the speed of the print head is equal to or higher than the reference speed. Done
The first setting is:
It is a setting to perform halftone processing using a dither mask with blue noise characteristics that can obtain a smooth print quality in which dots are well dispersed and dots are not noticeable,
The second setting is:
(A) When an image to be printed is yellow (RGB value, R = G = 255, B = 0-64, where RGB value is converted into 8 bits), the image is printed in each of the different states. The color difference in the CIE L * a * b * color space of the image is 2 or less as a change in b *,
(B) When the image to be printed is magenta (RGB value, R = B = 255, G = 0-64, where RGB value is converted into 8 bits), printing is performed in each of the different states. The color difference in the CIE L * a * b * color space of the resulting image is less than 0.5 as a change in L * or less than 0.5 as a change in a *, and
(C) When an image to be printed is cyan (RGB value, G = B = 255, R = 0-64, where RGB value is converted into 8 bits), printing is performed in each of the different states. The color difference in the CIE L * a * b * color space of the resulting image is less than 0.5 as a change in L *
A halftone process in which at least one of them is satisfied,
In the printing with the second setting, when printing is performed in each of a plurality of states in which the relative positions of the first pixel group and the second pixel group are different, images printed in the plurality of states are printed. A printing apparatus that suppresses the color difference to a predetermined value or less. The printing apparatus according to claim 1 ,
The plurality of states include a formation position of one dot belonging to the first pixel group and a formation position of a dot belonging to the second pixel group and corresponding to the one dot and adjacent to the one dot. The printing apparatus in which the difference in the distance between dots is 2/720 inch to 5/720 inch. The printing apparatus according to claim 1 or 2 ,
The print head includes dot forming elements arranged at a pitch corresponding to the print resolution in the width direction of the print medium,
The printing unit forms an image using the print head while transporting the print medium in a direction intersecting the width direction of the print head,
The first pixel group includes dots formed by dot forming elements arranged at predetermined positions in the transport direction of the print medium of the print head, and the second pixel group includes the dot of the print head. A printing apparatus including dots formed by dot forming elements arranged at a position different from the predetermined position in the conveyance direction of the print medium. The printing apparatus according to claim 1 or 2 ,
The print head is relatively movable in a main scanning direction that is a width direction of the print medium, and a plurality of nozzles that eject ink to the print medium are arranged in a direction intersecting the main scan direction. Comprising a plurality of nozzle rows spaced apart in the main scanning direction,
The printing unit repeatedly forms the image by repeating the main scanning and sub-scanning, which is a relative movement of the print medium in a direction crossing the scanning direction,
The first and second pixel groups have a relationship of any one of the following (A) and (C).
(A) The first pixel group includes dots formed when the print head moves forward in the main scanning direction, and the second pixel group includes dots formed when the print head moves backward in the main scanning direction. including.
(C) The first pixel group includes dots formed in one main scan of the print head, and the second pixel group is in a main scan different from the one main scan of the print head. Includes dots to be formed. A printed matter production method for producing printed matter by forming dots on a print medium using a print head,
The dot formation is performed at least in the first and second pixel groups, and at least a part of the dot formation by the first and second pixel groups is performed in the common region, so that the resolution N (N is 180 or more natural numbers) dpi (dot per inch), and color printing with multi-colored dots,
When forming dots on the print medium by the print head according to dot data representing the presence or absence of dot formation for each pixel, the dots are formed by dividing the dots into at least first and second pixel groups, An image is printed by performing at least a part of dot formation by the first and second pixel groups in the common area,
When the speed of the print head with respect to the print medium is less than a reference speed, printing is performed according to the first setting. When the speed of the print head is equal to or higher than the reference speed, printing is performed according to the second setting.
The first setting is:
It is a setting to perform halftone processing using a dither mask with blue noise characteristics that can obtain a smooth print quality in which dots are well dispersed and dots are not noticeable,
The second setting is:
(A) When an image to be printed is yellow (RGB value, R = G = 255, B = 0-64, where RGB value is converted into 8 bits), the image is printed in each of the different states. The color difference in the CIE L * a * b * color space of the image is 2 or less as a change in b *,
(B) When the image to be printed is magenta (RGB value, R = B = 255, G = 0-64, where RGB value is converted into 8 bits), printing is performed in each of the different states. The color difference in the CIE L * a * b * color space of the resulting image is less than 0.5 as a change in L * or less than 0.5 as a change in a *, and
(C) When an image to be printed is cyan (RGB value, G = B = 255, R = 0-64, where RGB value is converted into 8 bits), printing is performed in each of the different states. The color difference in the CIE L * a * b * color space of the resulting image is less than 0.5 as a change in L *
A setting that satisfies at least one of
In the printing with the second setting, when printing is performed in each of a plurality of states in which the relative positions of the first pixel group and the second pixel group are different, images printed in the plurality of states are printed. A printed matter production method that keeps the color difference below a predetermined value.

Images