JP2014076603A - Image recording device and image recording method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image recording device and an image recording method capable of suppressing density unevenness of each region without varying the number of recording dots and an arrangement form even in the case of a recording head having different recording characteristics in each region.SOLUTION: In order to make density among a plurality of regions uniform, from recording characteristics in each region, a distribution factor of large dots and small dots for each region is determined. Then, from the distribution factor and a quantization value, a distribution pattern that determines dots arrangement of the large dots and the small dots in each region is determined, and recording is performed according to the distribution pattern. At this time, in a plurality of distribution patterns corresponding to the equal quantization value, arrangement of pixels for recording the dots of any of the large dots and the small dots is fixed regardless of the distribution factor.

Description

  The present invention relates to an inkjet recording apparatus that records an image using a plurality of chips in which a plurality of nozzles that eject ink are arranged. In particular, the present invention relates to a recording method for correcting variations in recording characteristics among a plurality of chips.

  In an ink jet recording apparatus using a long recording head, recording characteristics such as ejection amount differ for each area of the recording head, and this variation in recording characteristics may appear as uneven density on the image. In particular, in a recording head configured by bonding a plurality of chips, recording characteristics are different for each chip, and density unevenness may be confirmed for each chip.

  Conventionally, for such density unevenness, a so-called head shading method in which the density between areas is adjusted by adjusting the number of dots recorded in a unit area has been employed. However, in such a head shading method, a visual discomfort associated with a difference in the number of dots between regions may be recognized.

  On the other hand, in Patent Document 1, a nozzle for ejecting large dots and a nozzle for ejecting small dots are prepared in advance on the recording head in order to correct the recording characteristic variation in the recording head, and the large dot for each region is prepared. And a technique for adjusting the ratio of ejecting small dots. With such a configuration, the density variation between areas can be corrected with the ratio of large dots to small dots while the number of dots recorded in each area is constant. No sense of incongruity due to the difference in the number of dots between them is recognized.

US Pat. No. 7,249,815

  However, in Patent Document 1, due to the configuration of the head, a position where large dots can be recorded and a position where small dots can be recorded are determined in advance. Therefore, the number of dots to be recorded on the paper surface can be unified between areas having different recording characteristics, but the arrangement state cannot be unified. For this reason, if there is a large variation in recording characteristics between areas, areas that are biased toward large dots and areas that are biased toward small dots appear, and the difference in dot patterns between these areas may be visually recognized. For this reason, in the case of adopting Patent Document 1, it is necessary to perform correction while maintaining a balance between image deterioration due to a difference in dot arrangement state and density unevenness due to a difference in recording characteristics. That is, the density correction can be performed only within a limited range where neither of the adverse effects is noticeable, and the density unevenness due to the difference in recording characteristics cannot be sufficiently suppressed.

  The present invention has been made to solve the above problems. Therefore, the purpose is to perform image recording that can suppress uneven density in each region without changing the number of dots to be recorded and the arrangement form, even if the recording head has different recording characteristics in each region. An apparatus and an image recording method are provided.

  Therefore, the present invention provides a recording head having a large dot nozzle array configured by arranging nozzles for recording large dots on a recording medium, and a small dot nozzle array configured by arranging nozzles for recording small dots. In an image recording apparatus for recording an image using a recording characteristic acquisition for acquiring recording characteristic information on the recording characteristics of the large dot nozzle array and the small dot nozzle array for dot recording in the same area of the recording medium for each area Means, a distribution rate determining means for determining a distribution ratio of large dots and small dots for each area based on the recording characteristic information, and a distribution pattern in which pixels for recording large dots and small dots in the area are determined. A plurality of distribution pattern storage means for storing a plurality of image data in association with a combination of the quantized value of the image data and the distribution rate, and the quantization value and the distribution rate, A distribution pattern setting unit that selects one of a plurality of distribution patterns and sets the distribution pattern for each region; and the large dots according to the distribution pattern set for each region by the distribution pattern setting unit A plurality of distribution patterns corresponding to the same quantized value among the plurality of distribution patterns, a large dot or a small dot, and a recording means for recording dots on a recording medium using the nozzle array and the small dot nozzle array The arrangement of the pixels for recording any one of the dots is constant regardless of the distribution rate.

  According to the present invention, even if a plurality of regions have different recording characteristics, the average discharge amount of each region is converged to a constant target discharge amount without changing the number of dots to be recorded and the arrangement form thereof, It is possible to reduce uneven density between regions.

FIG. 3 is a block diagram illustrating a configuration of image processing according to the first embodiment. 1 is a diagram illustrating a schematic configuration of a recording apparatus that can be used in the present invention. (A) And (b) is a detailed block diagram of a recording head. (A) And (b) is a flowchart explaining the processing process of CPU. (A) And (b) is a figure for demonstrating a multi-value error diffusion process. It is a schematic diagram for demonstrating concretely a mode that predetermined input image data is converted. It is the figure which showed the distribution pattern corresponding to a large and small dot distribution rate. (A) And (b) is the figure which showed the distribution pattern and the number of each large and small dots. It is a figure which shows the dot pattern generally used conventionally. (A) And (b) is a flowchart for creating a large and small dot distribution pattern. It is a figure which shows the dot pattern with respect to each level. It is a figure which shows a mode that a large and small dot distribution pattern is created. It is a figure which shows a mode that a dot is arrange | positioned based on a repulsive potential. (A) And (b) is a figure which shows the relationship between the distribution rate of large and small dots, and an average discharge amount. (A)-(c) is a figure explaining a mode that the average discharge amount is unified. It is a figure which compares the effect of this invention with the conventional head shading method. FIG. 10 is a block diagram illustrating a configuration of image processing according to a second embodiment. (A) And (b) is a flowchart explaining the processing process of CPU. It is a figure which shows the mode of recording and a reading part of a test pattern. It is a figure which shows a mode that data are distributed to each nozzle row based on a predetermined distribution rate. It is a figure which shows a mode that the same chip | tip is divided | segmented into a some area | region.

<Outline of line printer>
Hereinafter, embodiments of the present invention will be described in detail.

  FIG. 2 is a diagram showing a schematic configuration of a recording apparatus A1 that can be used in the present invention. The recording apparatus A1 is an ink jet line printer, and includes a control unit A2, ink cartridges A61 to A64, a recording head A7, a recording medium transport mechanism A8, and the like.

  The recording head A7 is a full-line type recording head, and has a plurality of thermal nozzles arranged in parallel in the X direction intersecting the transport direction (Y direction) on the surface facing the recording medium. Ink cartridges A61 to A64 store inks corresponding to cyan, magenta, yellow, and black, respectively, and are supplied to the individual nozzles of the recording head A7 through the ink introduction pipes A61a to A64a. From these nozzles, ink is ejected according to the image data, and recording is performed on the recording medium A100 that is transported at a constant speed in the Y direction. Details of the recording head A7 will be described later with reference to FIG.

  The recording medium transport mechanism A8 includes a paper feed motor A81 and a paper feed roller A82. The paper feed motor A81 rotates the paper feed roller A82 to transport the recording medium A100 to the recording head A7 at a constant speed in the Y direction.

  The control unit A2 is mainly composed of a CPU A3, a RAM A41, and a ROM A42, and processes received image data and performs a recording operation by controlling the recording head A7 and the paper feed roller A81. The CPU A3 performs various image processing as will be described later by expanding and executing the control program stored in the ROM A42 in the RAM A41. Then, image data that can be recorded by the recording head A7 is generated, and the recording head A7 and the recording medium transport mechanism A8 are controlled to record an image on the recording medium.

  3A and 3B are detailed configuration diagrams of the recording head A7. As shown in FIG. 3A, in the recording head A7 of this embodiment, a plurality of chips A71 to A74 in which a plurality of nozzle rows are arranged are continuous in the X direction while being alternately shifted in the Y direction. It is arranged. By ejecting ink from the individual nozzles of such a recording head onto a recording medium conveyed in the Y direction, an image corresponding to the recording head width w can be recorded.

  FIG. 3B is a diagram illustrating an arrangement state of nozzle rows in one chip A71. One chip A71 is provided with four nozzle rows A71a to A71d, and each nozzle row is composed of a plurality of nozzles arranged in a predetermined direction (here, the X direction) at a predetermined pitch. . Here, the nozzle arrays A71a and A71c are large dot nozzle arrays that record a large dot on a recording medium by ejecting a relatively large amount of ink droplets. On the other hand, A71b and A71d are small dot nozzle arrays that eject a relatively small amount of ink droplets to record small dots on a recording medium. In the recording medium transported in the Y direction, pixels on the same column extending in the Y direction are recorded alternately and selectively by the four nozzles included in these four nozzle rows.

<Image processing unit overview>
FIG. 1 is a block diagram showing the configuration of image processing according to this embodiment. Blocks A31 to A53 shown in FIG. 1 indicate individual functions of the control unit A2 of the recording apparatus A1 in FIG. FIGS. 4A and 4B are flowcharts for explaining the processing steps executed by the CPU A3 of this embodiment in association with the block diagram of FIG.

  FIG. 4A is a flowchart showing steps for the CPU A3 to determine the recording rate of the large dots and the small dots of each of the chips A71 to A74 in order to correct density unevenness. The process of FIG. 4A is performed prior to actual image recording. On the other hand, FIG. 4B is a flowchart of image processing performed by the CPU A3 when an actual image is recorded.

  First, referring to FIG. 1, a description will be given according to FIG. In step D01, the CPU A3 uses the recording characteristic acquisition unit A51 to obtain information on the average ejection amount of each of the four nozzle arrays arranged in the recording chips A71 to A74 (large dot recording characteristic information and small dot recording characteristic information). To get. In this embodiment, for such information, information determined at the time of shipment of the recording apparatus is stored in a ROM or the like, and the recording characteristic acquisition unit A51 reads it. Here, it is assumed that the average discharge amount of the large dot nozzle rows A71a and A71c is 3 ng and the average discharge amount of the small dot nozzle rows A71b and A71d is 2 ng.

  In subsequent step D02, CPU A3 sets a target discharge amount for each of chips A71 to A74 in correction target value setting unit A52. The target discharge amount is an ideal discharge amount that is commonly obtained for each chip in order to obtain a standard image density. In this embodiment, the target discharge amount of each chip is set to 2.5 ng.

  In step D03, the large / small dot distribution rate determination unit A53 sets the large dot recording rate and the small dot recording rate for each chip from the discharge amount information acquired in step D01 and the target discharge amount set in step D02. Stipulate That is, by adjusting the ratio (distribution rate) of the number of dots printed with large dots and the number of dots printed with small dots, the average ejection amount of the entire chip is set to 2.5 ng. In the case of the chip A71, since the average discharge amount of large dots is 3 ng, the average discharge amount of small dots is 2 ng, and the target discharge amount is 2.5 ng, the distribution ratio of large dots and small dots is 1: 1.

  As described above, when the large / small dot distribution ratio according to the ejection amount information is set for each chip, the present process is terminated.

  Next, with reference to FIG. 1, the image processing process during image recording will be described according to FIG. When the recording command is input, the CPU A3 reads the image data from the memory card A91 via the image input unit A31 and stores it in the RAM A41 (step D11). In this example, it is assumed that the resolution of the image data is 600 dpi (dot / inch), and each pixel is a color image of RGB each having 8 bits and 256 gradations. Of course, this is an example, and for example, a monochrome image can be similarly applied.

  In step D12, the CPU A3 performs color conversion processing of the image data in the color conversion processing unit A32. The color conversion process includes color correction, gradation correction, and color separation processes. The RGB (red, green, blue) luminance data of each pixel is converted into CMYK (corresponding to ink colors used by the printing apparatus). Cyan, magenta, yellow, and black) density data. As a result, 600 dpi 8-bit RGB data is converted to 600 dpi 8-bit CMYK data.

  In subsequent Step D13, the quantization processing unit A33 is used to perform quantization processing on the color-converted CMYK data. In this quantization process, 8-bit CMYK data represented by 256 values is converted into CMYK data represented by fewer level values (here, five values). As the quantization process, a dither method may be employed, but here, a multi-value error diffusion process is employed.

  FIGS. 5A and 5B are diagrams for explaining the multilevel error diffusion processing. In the figure, an input signal (In), which is 8-bit density data, is added with an accumulated error (dIn) generated from surrounding pixels by an adder 51 (In + dIn), and then a plurality of data prepared in advance by a comparator 52. Compared to the threshold value. FIG. 5B is a diagram showing a relationship between a plurality of threshold values compared by the comparator and an output value (Out) and an evaluation value (Ev) obtained as a result of the comparison.

  In the comparator 52, the input value (In + dIn) is quantized into five values of level 0 to level 4 according to the value. For example, when the input value (In + dIn) <32, the quantum value is level 0. When 96> (In + dIn)> 32, the quantum value is level 1. As described above, the comparator 52 changes the quantum value of any of level 0 to level 4 to the output value (in accordance with the magnitude relationship between the input value (In + dIn) and the four threshold values (32, 96, 160, 224). Output as Out). At this time, the comparator 52 also outputs an evaluation value Ev (0, 64, 128, 192, 255) corresponding to each quantum value to the subtractor 53.

  In addition to the evaluation value Ev from the comparator 52, the addition value (In + dIn) from the adder 51 is also input to the subtractor 53. The subtractor 53 calculates these differences, that is, Err = (In + dIn) −Ev as an error Err generated by quantization. Further, in order to distribute this to the surrounding pixels, a predetermined weighting operation is performed and added to the error buffer 54 prepared in association with the surrounding pixels. In the drawing, an example of a weighting coefficient when an error generated at the pixel position of the pixel of interest (*) is distributed to surrounding pixels in the error buffer 54 is shown. In the area corresponding to each pixel of the error buffer, a new error value is added every time the target pixel changes. When a new pixel of interest is quantized, the accumulated error at the corresponding pixel position is read from the error buffer 54 and normalized by the sum of the weighting coefficients, and is sent to the adder 51 as the accumulated error dIn. To do.

  As described above, in the multi-value error diffusion processing employed in the present embodiment, the 256-gradation density data of each pixel is converted into the 5-value density data while the error generated in each pixel is distributed to the peripheral pixels. Convert to

  Returning to FIG. When the quantization process D13 is completed, the CPU A3 inputs the 600 dpi 5-value quantum value to the dot distribution pattern setting unit A34, and determines the arrangement of large and small dots in the target pixel (D14). Referring to FIG. 1, large and small dot distribution patterns corresponding to various distribution rates are prepared for each level (1 to 4) in the large and small dot distribution pattern storage unit A41. The dot distribution pattern setting unit A34 is based on the quantum value obtained from the quantization processing unit A33 and the distribution rate obtained from the large / small dot distribution rate determination unit A53, and the distribution pattern stored in the large / small dot distribution pattern storage unit A41. Select one of the following. As a result, the quantum value of 600 dpi 5 value is converted into 1200 dpi binary data for large dots and 1200 dpi binary data for small dots. The distribution patterns stored in the large / small dot distribution pattern storage unit A41 will be described in detail later.

  In step D15, the CPU A3 uses the used nozzle row determination unit A36 to distribute each of the large dot data and the small dot data to a plurality of nozzle rows on the chip. Specifically, a mask pattern stored in advance in the nozzle array distribution pattern portion A42 is read out, and a logical product between the mask pattern and the large dot data or small dot data determined by the dot distribution pattern setting portion A34 is obtained. Perform the operation. As a result, the dot pattern actually recorded by each nozzle row is determined. The mask pattern will be described in detail later. The recording data as the determined dot pattern is transferred to a buffer to be recorded by each nozzle row for each chip on the recording head.

  In step D16, an image is recorded by ejecting ink from each nozzle in accordance with the recording data determined for each nozzle row. That is, the CPU A3 drives the paper feed motor A81, and ejects ink from the recording head A7 based on the nozzle row-specific recording data in accordance with this movement. As a result, large dots and small dots are recorded at an appropriate ratio for each chip, and ink droplets corresponding to the target ejection amount of 2.5 ng are ejected from all the chips, and an image having a uniform density is output. This process is completed.

  FIG. 6 is a schematic diagram for specifically explaining how predetermined input image data is converted by the above-described steps.

  Reference numeral 601 denotes input image data received by the image input unit A31, which indicates a 600 dpi 4 × 4 pixel region. In this example, each pixel has data of (R, G, B) = (192, 192, 192). Reference numeral 602 denotes cyan multivalued data after the color conversion processing unit A32 performs color conversion on the data of each pixel in the image area. Here, the signal value of cyan multilevel data is C = 64. Reference numeral 603 denotes a result obtained by quantizing the multilevel data 602 by the quantization processing unit A33 according to the multilevel error diffusion method described with reference to FIGS. 5A and 5B. ) Is quantized. Reference numeral 604 denotes a dot pattern selected by the dot distribution pattern setting unit A34 from the distribution patterns stored in the large and small dot distribution pattern storage unit A41 based on the predetermined distribution ratio (1: 1) and the quantization value 603. .

  Here, the distribution pattern stored in the large / small dot distribution pattern storage unit A41 will be described.

  FIG. 7 is a diagram showing distribution patterns corresponding to various large and small dot distribution ratios when the quantization value is level 1 (L1). The 600 dpi 4 × 4 pixel area 701 having level 1 is converted into any of the 1200 dpi 8 × 8 pixel areas 702 to 706 in which large dots and small dots are recorded or not recorded, depending on the distribution ratio. Is done. Here, when 702 is a large / small dot distribution ratio of 1: 0, 703 is 3: 1, 704 is 1: 1, 705 is 1: 3, 706 is The case of 0: 1 is shown respectively. In the figure, ◎ indicates a pixel that records a large dot, ◯ indicates a pixel that records a small dot, and a blank indicates a pixel that does not record any dot. When these four patterns 702 to 706 are compared, the number of large dots and the number of small dots are different in each pattern, but the positions where dots are recorded are unified at any distribution ratio. In the present embodiment, such a feature is common to all levels. That is, areas having the same level are recorded with the same dot pattern even if the recording characteristics are different from each other, and discontinuity due to the difference in the dot pattern is not confirmed.

  FIGS. 8A and 8B are diagrams showing a distribution pattern when the large / small dot distribution ratio is 1: 1, and the numbers of large and small dots obtained on the recording medium. FIG. 8A is a diagram showing an actual distribution pattern for each of level 1 to level 4 when the large / small dot distribution ratio is 1: 1. A 600 dpi 4 × 4 pixel region 801 having level 1 (L1) is converted into a 1200 dpi 8 × 8 pixel region 802. A 600 dpi 4 × 4 pixel region 803 having level 2 is converted into a 1200 dpi 8 × 8 pixel region 804, and a 600 dpi 4 × 4 pixel region 805 having level 3 is converted into a 1200 dpi 8 × 8 pixel region 806. Also, the 600 dpi 4 × 4 pixel region 807 having level 4 is converted into a 1200 dpi 8 × 8 pixel region 808. In any dot pattern, the number of large dots and the number of small dots are the same number (1: 1).

  FIG. 8B is a diagram showing the number of large dots and small dots actually recorded for the output value from the quantization processing unit. As the number of levels increases, the number of large dots and the number of small dots also increase, but the ratio is kept at 1: 1.

  In this embodiment, a dot pattern is recorded so that dots are recorded even at higher levels (for example, dot recording pixels of pattern 804) in pixels where dots are recorded at lower levels (for example, dot recording pixels of pattern 802). The ordering rules are maintained. For this reason, even in a uniform image whose level gradually changes, discontinuity due to an increase in dots is not confirmed, and a smooth image can be recorded.

  In the present embodiment, a plurality of large and small dot distribution patterns prepared in association with each quantum value and each distribution rate are stored in the large and small dot distribution pattern storage unit A41.

  Returning to FIG. The large / small dot distribution pattern 604 is distributed to a large dot pattern 605 and a small dot pattern 604. Thereafter, the used nozzle row determination unit A36 uses the mask patterns 607 and 608 stored in the nozzle row distribution pattern storage unit A42 to change the large dot pattern 605 and the small dot pattern 606 to the nozzle rows A71a to A71d, respectively. Distribute.

  This will be specifically described. In the mask patterns 607 and 608, the pixels shown in gray are pixels that allow dot recording, and the pixels shown in white are pixels that do not allow dot recording. The used nozzle array determination unit A36 obtains dot data 609 to be recorded by the large dot nozzle array A71a by performing a logical product operation of the large dot pattern 605 and the mask pattern 607. Further, by performing a logical product operation of the large dot pattern 605 and the mask pattern 608, dot data 610 to be recorded by the large dot nozzle row A71c is obtained. Further, the used nozzle row determination unit A36 obtains dot data 611 to be recorded by the small dot nozzle row A71b by calculating the small dot pattern 606 and the mask pattern 607. Further, the dot data 612 to be recorded by the small dot nozzle array A71d is obtained by calculating the small dot pattern 606 and the mask pattern 608.

  Thereafter, dots are recorded according to the dot data 609 to 612 corresponding to each nozzle row, whereby a dot pattern 613 is obtained on the recording medium.

  As described above, according to this embodiment, even if a plurality of chips have different recording characteristics, the average discharge amount of each chip is converged to a constant target discharge amount, and uneven density between chips is reduced. I can do it.

  Conventionally, the quantized data after the multi-value quantization processing is generally converted into binary data using a dot pattern without distinction between large and small dots as shown in FIG. Therefore, when the distribution ratio of large and small dots is adjusted for each chip as in the present invention, after performing binarization processing using such a conventional dot pattern, individual distribution is performed according to the corresponding distribution ratio. It is also possible to distribute the dots into large dots and small dots. However, in this case, a two-stage process and configuration of a binarization process according to a conventional dot pattern and a distribution process for large dots and small dots are required.

  On the other hand, the dot distribution pattern setting unit A34 of this embodiment can uniquely set the large and small dot distribution patterns in which the arrangement of the large dots and the small dots is already determined in association with the quantum value and the distribution rate. ing. That is, binarization processing and distribution processing to large dots and small dots are performed in a lump, and the processing load and processing time can be reduced compared to the case of using the conventional dot pattern. .

<Generation method of large and small dot distribution pattern>
Hereinafter, an example of a process for creating a large / small dot distribution pattern stored in the large / small dot distribution pattern storage unit A41 of the present embodiment will be described.

  FIGS. 10A and 10B are flowcharts for explaining an example of a process for creating a large and small dot distribution pattern. FIG. 10A shows a method of creating using a random number, and FIG. 10B shows a method of creating using a repulsive potential.

  In FIG. 10A, first, in step N00, a dot pattern of a quantization level X that is a target for generating a large and small dot distribution pattern is acquired.

  FIG. 11 is a diagram showing a dot pattern for each level. The dot patterns 11a to 11d correspond to level 1 to level 4, respectively. These dot patterns are predetermined patterns, for example, by cyclically arranging the conventional dot patterns as shown in FIG. 9, and dot recording (1) and non-recording (0) for each pixel. Only. In step N00, X = 1 is initially set, and a level 1 dot pattern 11a is acquired.

  In the subsequent step N01, for each distribution rate, the already created large / small dot distribution pattern of level (X-1) is set as the current large / small dot distribution pattern. When X = 1, a large and small dot distribution pattern of level 0 is set, but this pattern is a common pattern for all distribution ratios in which no dot is arranged in any pixel.

  In Step N02, the large dot generation establishment “Pro_L” corresponding to each distribution rate is calculated. For example, when the distribution ratio of large: small is 3: 1, Pro_L = 3 / (3 + 1) × 100 = 75 (%). Further, when the large: small distribution ratio is 1: 1, Pro_L = 1 / (1 + 1) × 100 = 50 (%).

  In step N03, one of the dots in which the dots are not arranged in the current large / small dot distribution pattern set in step N01 is selected from the recording dots of the dot pattern acquired in step N00 and set as the target dot. .

  In the subsequent step N04, an integer rand between 1 and 100 is generated as a random number, and the generated random number rand is compared with the large dot generation probability “Pro_L” calculated in step N02. If rand> Pro_L, the process proceeds to step N06 to distribute the target dot to small dots. On the other hand, if rand ≦ Pro_L, the process proceeds to step N07 to distribute the target dot to large dots. In this way, the large / small dot distribution pattern is updated.

  In Step N08, it is determined whether or not there is a dot that has not been distributed to any of the large and small dots in the dot pattern acquired in Step N00. If it is determined that there is a dot that has not yet been distributed, the process returns to step N03 to distribute the next dot. On the other hand, if it is determined that all the dots have been distributed, the process proceeds to step N09 to determine the current large / small dot distribution pattern as the level X large / small dot distribution pattern. The processes from step N03 to N09 are performed for each distribution rate, and as many large and small dot distribution patterns as the number of distribution rates are prepared.

  In Step N10, it is determined whether or not the setting of the large / small dot distribution pattern has been completed for all levels. If the level to be set still remains, the process proceeds to step N11, and after incrementing X, the process returns to step N00 to create the next-level large / small dot distribution pattern. On the other hand, if it is determined in step N10 that the setting of the large / small dot distribution pattern has been completed for all levels, this processing is terminated.

  In this way, in this example, the level X dot pattern is created based on the level (X-1) dot pattern. That is, a pixel that records a large dot with a level (X-1) dot pattern records a large dot with a level X dot pattern, and a pixel that records a small dot with a level (X-1) dot pattern Even small dot patterns are designed to record small dots. Therefore, even when a gradation image whose level gradually changes is recorded, discontinuity due to the increase in dots is not confirmed, and a smooth image can be output.

  On the other hand, FIG. 10B is a flowchart for explaining a method of creating a large and small dot distribution pattern using a repulsive potential. FIG. 12 is a diagram showing how large and small dot distribution patterns are created according to this flowchart.

  In FIG. 10B, Steps N20 and N21 are the same as Steps N00 and N01 in FIG. In Step N20, X = 1 is initially set, and the level 1 dot pattern 101 in FIG. 12 is acquired.

  In Step N22, the required large dot number ND at level X is calculated from each distribution rate. For example, at level 1, the number of recording dots included in the dot pattern 101 acquired at step N20 is 16 dots in the 8 × 8 pixel area. Therefore, when the distribution ratio is 1: 1, the required large number of dots ND is ND = 16 × 1 / (1 + 1) = 8 dots.

  In step N23, the “repulsive potential_integrated value” is the highest among the dots in which the dots are not arranged in the current large / small dot distribution pattern set in step N21 among the recorded dots of the dot pattern obtained in step N20. Select a smaller position. Regarding the distribution of the first dot of level 1, since “repulsive potential_integrated value” is “0” at any position, an arbitrary dot position is selected. Here, it is assumed that a position indicated by an asterisk in the pattern 102 is selected.

  In step N24, the dots selected in step N23 are distributed to large dots. A pattern 103 in FIG. 12 shows a state in which the dots are distributed into large dots (）).

  In step N25, the repulsive potential of the distributed large dots is added to “repulsive potential_integrated value”. The repulsive potential will be described below.

  FIG. 13 is a diagram showing a repulsive potential in this embodiment and how dots are arranged based on the repulsive potential. In this embodiment, a repulsive potential having a large inclination is generated around the arranged large dots. Specifically, when a large dot is placed on one pixel, the potential of the pixel is “50000”, and the potential of the other pixels is “10000 / the fourth power of the distance to the large dot arrangement pixel”.

  A1 in FIG. 13 is a bird's-eye view showing the repulsive potential generated by arranging a large dot at coordinates (4, 4) as a value in the z direction with respect to the xy plane. Further, a2 is a table showing the value of the repulsive potential at each coordinate (x, y).

  On the other hand, b1 in FIG. 13 is a diagram showing a repulsive potential generated by arranging a large dot at the origin coordinates (0, 0) in a contour graph. Further, b2 is a table showing the value of the repulsive potential at each coordinate (x, y) in this case. It can be seen that the state of the coordinates (4, 4) at a1 and a2 has moved to the origin (0, 0).

Here, if the repulsive potential of a single dot is Pot_alone, the potential at position (x, y) is
Pot_alone = 50000 (x = 0, y = 0)
10000 ÷ (x ^ 2 + y ^ 2) ^ 2 (x ≠ 0, y ≠ 0)
It becomes. Assuming that the same 8 × 8 pattern continues on the right, lower right, and lower side in order to satisfy the boundary condition, the repulsive potential Pot (x, y) at (x, y) is Pot_0 (x, y) = Pot_alone. (X, y)
+ Pot_alone (x + array_X, y)
+ Pot_alone (x-array_X, y)
+ Pot_alone (x, y-array_Y)
+ Pot_alone (x + array_X, y-array_Y)
+ Pot_alone (x-array_X, y-array_Y)
+ Pot_alone (x, y + arlya_Y)
+ Pot_alone (x + array_X, y + array_Y)
+ Pot_alone (x-array_X, y + array_Y)


array_X: number of pixels in the x direction of the dot pattern (8 in this embodiment)
array_Y: number of pixels in the y direction of the dot pattern (8 in this embodiment)
It becomes.

  C1 and c2 in FIG. 13 show the state of repulsive potential when the same pattern is continued on the right, lower right, and lower sides in order to satisfy the boundary condition with respect to b1 and b2.

Here, the repulsive potential of (x, y) when a large dot is arranged at an arbitrary position (a, b) is obtained by substituting the relative position from (a, b) into the above Pot_0 (x, y). Therefore, the repulsive potential is Pot_ab (x, y) = Pot_0 (Pos_x, Pos_y)
Pos_x = x−a (x ≧ a)
a−x (x ≦ a)
Pos_y = y−b (y ≧ b)
by (y ≦ b)
It becomes.

  Returning to FIG. 10B again. In step N23, the coordinate having the smallest value of the repulsive potential is selected in the 8 × 8 pixel region while referring to the repulsive potential space, and in step N24, a large dot is arranged at the coordinate. Further, in step N25, the repulsive potential integrated value is calculated again according to the above equation with the arrangement of a new large dot.

  Here, to explain again along the example of FIG. 12, when the large dot is arranged at the position (x, y) = (7, 4) indicated by (◎) of the pattern 103 in step N24, each pixel is changed. The repulsive potential is as shown in Table 107 and the contour line graph 108. In Table 107, shaded pixels indicate pixels in the pattern 101 where dots are present.

  In the subsequent step N26, the number of arranged large dots D is incremented by 1 (D = D + 1) in accordance with the arrangement of large dots in step N24.

  In Step N27, it is determined whether or not the arranged large dot number D has reached the necessary large dot number ND calculated in Step N22. If D ≠ ND, the process returns to step N23 to place the next large dot. Here, in order to arrange the second large dot, the description will be continued by returning to Step N23 again.

  In Step N23, the position where the “repulsive potential_integrated value” is the smallest is selected from the dots where the dots are not arranged in the current large / small dot distribution pattern among the recording dots of the dot pattern acquired in Step N20. . As already described, in the table 107 showing the repulsive potential when single dots are arranged, the pixels where the dots are present are shaded. Therefore, in step N23 at this stage, a pixel having the smallest “repulsive potential_integrated value” is selected from the pixels indicated by shading in Table 107. If there are a plurality of such pixels, a random number is generated to select one pixel. Here, it is assumed that (x, y) = (2, 7) is selected from (2, 2) and (2, 7). In this case, in the subsequent step N24, a large dot is newly arranged at coordinates (x, y) = (2, 7), and the pattern 104 is obtained. Further, in step N25, a new integrated value of the famous power potential is obtained as shown in Table 109 and the contour line graph 110.

  Thereafter, when the number of arranged large dots D is further incremented by 1 in step N26, it is determined again in step N27 whether the number of arranged large dots D has reached the required number of large dots.

  As described above, the arrangement of the large dots in the pixel having the lower repulsive potential and the recalculation of the repulsive potential integrated value are repeated until the number D of arranged large dots reaches the required large number of dots ND, and then the process proceeds to Step N28. move on. A pattern 105 in FIG. 12 shows a dot pattern when the number D of arranged large dots becomes equal to the required number of large dots ND (D = ND = 8).

  In Step N28, all the remaining dots other than those distributed to the large dots among the recording dots of the dot pattern acquired in Step N20 are distributed to the small dots. The pattern 106 in FIG. 12 shows a state in which the remaining 8 dots are distributed to small dots and all 16 dots are distributed to 8 large dots and 8 small dots.

  In step N29, the completed large / small dot distribution pattern is determined to be a level X large / small dot distribution pattern. In the above description, the dot pattern when the distribution ratio is 1: 1 has been described as an example. However, the above steps N23 to N29 are performed for each distribution ratio, and the large and small dot distribution patterns are distributed. Only the number of rates are prepared.

  In step N30, it is determined whether the setting of the large / small dot distribution pattern has been completed for all levels. If there is still a level to be set, the process proceeds to step N31. After incrementing X, the process returns to step N20 to create the next-level large / small dot distribution pattern. On the other hand, if it is determined in step N20 that the setting of the large / small dot distribution pattern has been completed for all levels, this processing is terminated.

  As described above, in this example as well, as in FIG. 10A, the level X dot pattern is created with reference to the level (X-1) dot pattern, thereby avoiding discontinuities due to the increase in dots. I can do it. In addition, as in this example, the method of arranging large dots at a position where the repulsive potential is as low as possible is larger in the large and small dot distribution pattern than the method of using random numbers described in FIG. The dispersibility of dots can be further increased. As a result, low frequency components of the spatial frequency of large dots that are relatively easy to visually recognize are suppressed, and good results are obtained with respect to graininess and uniformity.

<Effect of the present invention>
Hereinafter, the effect of the present embodiment will be described.

  FIGS. 14A and 14B are diagrams showing how the average discharge amount is adjusted by changing the distribution ratio between large dots and small dots in one chip. In this embodiment, as shown in FIG. 14A, the recording rate of each dot is made variable while maintaining the sum of the recording rate of large dots and the recording rate of small dots in one chip at 100%. Yes. That is, when the large dot recording rate is 100%, the small dot recording rate is 0%, and when the large dot recording rate is 0%, the small dot recording rate is 100%.

  When the recording rate of large dots and small dots, that is, the distribution rate is gradually changed while maintaining such a relationship, the average ejection amount of the chip gradually changes as shown in FIG. For example, in a chip having a large dot discharge amount of 3 pl and a small dot discharge amount of 2 pl, when the large dot recording rate is 100% and the small dot recording rate is 0%, the average discharge amount AV is AV = ( 100 × 3 + 0 × 2) / 100 = 3 pl. When the large dot recording rate is 0% and the small dot recording rate is 100%, the average discharge amount AV is AV = (0 × 3 + 100 × 2) / 100 = 2 pl. Further, when the recording rate of large dots is 50% and the recording rate of small dots is 50%, the average discharge amount AV is AV = (50 × 3 + 50 × 2) /100=2.5 pl. That is, in a chip with a large dot discharge amount of 3 pl and a small dot discharge amount of 2 pl, the average discharge amount of the chip is changed between 2 pl and 3 pl by adjusting the distribution ratio of the large dots and the small dots. I can do it. In other words, even with a plurality of chips having different discharge amounts, the average discharge amount of all the chips can be made substantially constant by adjusting the distribution ratio of the large dots and the small dots with each chip.

  FIGS. 15A to 15C are views for explaining a state in which the average discharge amount is unified among a plurality of chips having different discharge amount errors. FIG. 15A is a graph and a table showing the relationship between the ejection amount error rate and the ejection amount of large and small dots in each chip. Here, when manufacturing a chip having large and small nozzles with a large dot target discharge amount of 3 pl and a small dot target discharge amount of 2 pl, a discharge amount error of ± 20% occurs during manufacturing. . If the chip has a discharge amount error rate of 0%, the discharge amount of large dots is 3 pl, the discharge amount of small dots is 2 pl, and the average discharge amount is 2.5 pl. If the chip has a discharge amount error rate of + 20%, the discharge amount of large dots is 3.6 pl, the discharge amount of small dots is 2.4 pl, and the average discharge amount is 3 pl. If the chip has a discharge amount error rate of −20%, the discharge amount of large dots is 2.4 pl, the discharge amount of small dots is 1.6 pl, and the average discharge amount is 2 pl.

  FIG. 15B is a diagram showing a state in which the present embodiment is adopted to adjust the usage ratio, that is, the printing rate of each of the large dots and the small dots in accordance with the ejection amount error rate. If the chip has a discharge rate error rate of 0%, both the large dot recording rate and the small dot recording rate are set to 50%. If the chip has a discharge rate error rate of + 20%, the large dot recording rate is set to 0%, and the small dot recording rate is set to 100%. If the chip has a discharge rate error rate of −20%, the large dot recording rate is set to 100% and the small dot recording rate is set to 0%. In the case of other discharge amount error rates between −20% and + 20%, the recording rate is obtained by changing the above values stepwise as shown in the figure.

  FIG. 15C is a diagram showing an average discharge amount when the recording rate is set as shown in FIG. When the ejection amount error rate is 0%, both the recording rate for large dots and the recording rate for small dots are 50%, so the average ejection amount is 0.5 × 3 + 0.5 × 2 = 2.5 pl. When the ejection amount error rate is + 20%, the recording rate for large dots is 0% and the recording rate for small dots is 100%, so the average ejection amount is 0 × 3.6 + 1 × 2.5 = 2.5 pl. . When the discharge rate error rate is −20%, the recording rate for large dots is 100% and the recording rate for small dots is 0%, so the average discharge rate is 1 × 2.5 + 0 × 1.6 = 2.5 pl. Become. That is, if a recording rate corresponding to the discharge rate error rate is set as shown in FIG. 15B, the average discharge rate is unified to 2.5 pl even for a plurality of chips having different discharge rate error rates. I can do it.

  In FIG. 15C, when the recording rate is set as shown in FIG. 15B and when the recording rate is not set (that is, both the recording rate of large dots and the recording rate of small dots are fixed to 50%). ) And the average discharge amount. It can be seen that by adopting the configuration of this embodiment, a uniform average discharge amount can be obtained regardless of the discharge amount error.

  Here, the recording rate of the large dots and the small dots can be adjusted in the range of 0 to 100%. However, from the viewpoint of the life of the head and the like, the ejection frequency is largely biased to either the large dots or the small dots. It may not be preferable. In such a case, the variable range of the recording rate, that is, the adjustment range of the distribution rate of large and small dots may be made narrower.

  On the other hand, FIG. 16 is a diagram comparing the effect of this embodiment with a conventional head shading method. Patterns 16a to 16c show dot recording states when the discharge amount variation is adjusted by the head shading method, that is, by the number of dots to be recorded in the unit area. Here, the target ejection amount is 2.5 pl, the pattern 16a is a dot arrangement state when the ejection amount error rate is −20%, the pattern 16b is a dot arrangement state when the ejection amount error rate is 0%, and the pattern 16c is ejection. The dot arrangement state when the amount error rate is + 20% is shown. In the pattern 16b with a discharge amount error rate of 0%, 16 dots are recorded in the 8 × 8 pixel area. With this as a reference, in the pattern 16a, 16 ÷ 0.8 = 20 dots are recorded in order to compensate for a shortage of -20% in the discharge amount. Further, in the pattern 16c, 16 ÷ 1.2≈13 dots are recorded in order to correct + 20% excess discharge amount.

  On the other hand, the patterns 16d to 16f show dot recording states when the variation in the ejection amount is adjusted according to this embodiment. Pattern 16d is a dot arrangement state when the discharge amount error rate is −20%, pattern 16e is a dot arrangement state when the discharge amount error rate is 0%, and pattern 16f is a dot arrangement when the discharge amount error rate is + 20%. Each state is shown. In the patterns 16d to 16f, the large dots indicate the recording positions of the small dots, and in the pattern 16e having a discharge amount error rate of 0%, the large dots are 8 and the small dots are 8 in the 8 × 8 pixel region. A total of 16 dots are recorded. On the other hand, in the pattern 16d, 16 dots of large dots are recorded to make up for the shortage of -20%, and in the pattern 16f, 16 dots of small dots are recorded to correct the excess of + 20%.

  Comparing these two adjustment methods, in either method, the dot coverage per unit area and thus the optical density can be stabilized between the patterns 16a to 16c or between the patterns 16d to 16f regardless of the ejection amount error rate. I can expect. However, in the head shading method, as seen in the patterns 16a to 16c, the number of dots to be recorded and the dot arrangement are different depending on the ejection amount error rate, and the difference in the dot arrangement state gives an impression of being visually uneven. May give. On the other hand, in the method of the present embodiment, as seen in the patterns 16d to 16f, although the large dots and the small dots are interchanged, the number of dots to be recorded and the arrangement thereof are not changed, and visually uneven. It's hard to give an impression.

  That is, according to the present embodiment, even when a recording head in which a plurality of chips having different ejection amounts are arranged is used, density unevenness can be achieved without differentiating the number of dots to be recorded in the unit area and the arrangement state between the chips. It is possible to record a uniform image with no image.

  Also in this embodiment, the recording apparatus and the recording head described with reference to FIGS. 2 and 3 are used.

  FIG. 17 is a block diagram illustrating a configuration of image processing according to the present exemplary embodiment. The difference from the first embodiment described in FIG. 1 is that the function of the dot distribution pattern setting unit A34 and the function of the used nozzle row determination unit A36 in the first embodiment are combined into one nozzle row-specific dot distribution pattern setting unit A341. It is that.

  FIGS. 18A and 18B are flowcharts for explaining the processing steps executed by the CPU A3 of this embodiment in association with the block diagram of FIG. FIG. 18A is a flowchart for setting the recording characteristic information of each chip, and the process is the same as FIG. 4A described in the first embodiment. However, in step D01 of the first embodiment, the recording characteristic information of each chip stored in the ROM is read. However, in step S01 of this embodiment, the image reading means reads the test pattern actually recorded by each chip. The recording characteristics are set from the brightness information obtained by reading.

  FIG. 19 is a diagram showing how the test pattern recording and reading unit J1 reads this in this embodiment. An image reading unit J1 is provided downstream of the recording head A7 in the Y direction, and reads the brightness of the test pattern J100 recorded by the recording head A7. Here, the image reading unit J1 is composed of a plurality of CCDs arranged in the X direction, and the CPU A3 obtains the lightness distribution of the test pattern J100 in the X direction based on the lightness information obtained from each of the plurality of CCDs. The recording characteristic information for each chip is defined.

  On the other hand, FIG. 18B is a flowchart of image processing performed by the CPU A3 of this embodiment when an actual image is recorded. In this flowchart, steps S11 to S13 and step S15 are the same as steps D11 to D13 and step D16 in FIG.

  FIG. 20 is a diagram illustrating a state in which the CPU A3 distributes data to each nozzle row based on a predetermined distribution rate in step S14. Here, when the distribution ratio is 1: 1, how the quantized values of level 1 to level 4 correspond to the two large dot nozzle rows A71a and A71c and the two small dot nozzle rows A71b and A71d. Indicates whether dots (recording data) are distributed. In the present embodiment, the large and small dot distribution pattern storage unit A41 in FIG. Is remembered. Then, the dot distribution pattern setting unit A341 for each nozzle row accesses the large / small dot distribution pattern storage unit A41 based on the quantum value obtained from the quantization processing unit A33 and the distribution rate obtained from the large / small dot distribution rate determination unit A53. . As a result, a distribution pattern corresponding to each of the four nozzle arrays A71a to A71d is selected. That is, the 600 dpi 5-value quantum value is converted into 1200 dpi 2-value data for the large dot nozzle rows A71a and A71c and 1200 dpi 2-value data for the small dot nozzle rows A71b and A71d. As described above, the large and small dot distribution pattern storage unit A41 of this embodiment has various distribution patterns for the four nozzle rows corresponding to the quantized values (level 1 to level 4) as shown in FIG. It is prepared according to the distribution rate.

  According to the present embodiment described above, mask patterns 607 and 608 are prepared in addition to the large and small dot distribution patterns as in the first embodiment, or a logical product operation is performed between the mask pattern and the quantized data. There is no need to do. Therefore, the arithmetic circuit is simplified and the processing time can be shortened as compared with the first embodiment.

  The dot pattern of the second embodiment shown in FIG. 20 uses the same method as that of the first embodiment described with reference to FIG. 10, that is, the “method for generating random numbers” and the “method using repulsive potential”. Can be created and each effect can be obtained. Furthermore, when creating a dot pattern using repulsive potential, by arranging the dots while minimizing the integrated value of the repulsive potential for each nozzle array, the dot dispersion of individual nozzle arrays as well as large and small dots Can also be increased. As a result, the life of the recording head can be stabilized without biasing the frequency of use of the individual nozzle rows. In this case, the density unevenness due to the recording position shift for each nozzle row can also be reduced.

  In the above description, it is assumed that the discharge rate error rate differs for each chip, and the distribution ratio of large and small dots is set for each chip. However, even in the same chip, there may be large variations in recording characteristics. is there. In such a case, for example, as shown in FIG. 21, a plurality of nozzles arranged on the same chip A71 may be divided into a plurality of regions, and the distribution ratio of large and small dots may be set for each region. FIG. 21 shows a case where the area on the chip is divided into three.

  Further, in the first embodiment, the discharge amount information stored in the ROM and the lightness information of the test pattern acquired by the image reading unit in the second embodiment are described as the recording characteristic information. However, it is not limited to these. In the present invention, the recording characteristic information may be a parameter that serves as a guideline for determining the distribution ratio of large dots and small dots in order to adjust the density of the recording medium as a result. For example, the average discharge amount of each chip is classified into several ranks, and rank information can be stored in the ROM instead of the discharge amount information of the first embodiment when manufacturing the printing apparatus. In this case, memory management and subsequent processing can be simplified as compared with the first embodiment. Further, by utilizing the fact that the ejection amount correlates with the size of the ejection port, the diameter of the ejection port of each chip can be measured at the time of manufacturing the recording apparatus, and the value can be managed as recording characteristic information.

  Further, when recording and detecting a test pattern as in the second embodiment, for example, an area corresponding to a part of each chip is detected instead of the entire area of the test pattern, and the obtained recording characteristic information is obtained. The recording characteristic information of the entire corresponding chip may be used. In this way, it is possible to reduce the ink consumption and the processing time for acquiring the recording characteristic information, and this is particularly effective when the recording characteristic information varies in units of chips. Further, density information may be measured instead of the lightness information of the second embodiment and managed as recording characteristic information.

  In addition, the image reading means does not read the recorded test pattern, but the user visually confirms the test pattern and inputs the density distribution to the recording device to acquire the recording characteristic information of each region. You can also. Even if there is a variation in the discharge amount between chips, how much the variation is noticeable on the image varies depending on the type of the recording medium. By making the recording characteristic information depend on the visual judgment of the user, unnecessary correction can be avoided.

  In the above embodiment, the recording apparatus has all the functions shown in FIG. 1 and FIG. 17, but part or all of the processing steps shown in FIG. 4 and FIG. 18 are connected to the recording apparatus. It can also be performed by a host device that has been selected. In this case, the entire system including the host device and the recording device is the image recording device of the present invention.

  Further, in the above embodiment, the discharge amount error rate is determined for each chip, and it has been described on the assumption that the large dots and the small dots recorded on the same chip (or region) have the same discharge amount error rate (FIG. 15). However, the large dot nozzle array and the small dot nozzle array may be formed on different chips, and the large dot and the small dot that record the same area of the recording medium do not always have the same discharge amount error rate. In such a case, the recording characteristics for large dots and the recording characteristics for small dots are managed independently, and the combination of the large dot recording characteristics and the small dot recording characteristics for recording the same area of the recording medium. What is necessary is just to set the distribution rate between dots.

  Furthermore, in the above-described embodiment, the same area of the recording medium is recorded by two large dot nozzle arrays and two small dot nozzle arrays. However, the present invention is not limited to this. There may be one large dot nozzle row and one small dot nozzle row, or three or more rows. Also, the large dot nozzle row and the small dot nozzle row may be different numbers of nozzle rows. Furthermore, the types of dots that record the same area may extend to a larger number of sizes, such as large dots, medium dots, and small dots. In any case, it is within the scope of the present invention as long as the recording rate of each dot of a plurality of sizes is appropriately adjusted for each region and the recording operation is performed based on the characteristic distribution pattern of the present invention.

  Furthermore, in the above embodiment, a line printer as shown in FIG. 1 is used as an example of using a long recording head having a plurality of chips. However, a recording head having a plurality of chips or having different recording characteristics for each region is not limited to a full-line type recording head. In recent years, a long recording head is sometimes used even in a serial recording apparatus. Even in such a form, the above-described embodiment can be applied to obtain the effect of the present invention.

A1 recorder
A2 control unit
A3 CPU
A7 Recording head
A34 Dot distribution pattern setting section
A36 Used nozzle row determination unit
A41 Large and small dot distribution pattern storage
A42 Nozzle row distribution pattern storage unit
A51 Recording characteristics acquisition unit
A53 Large / small dot distribution ratio determining unit A71 to A74 Chips A71a, A71c Large dot nozzle rows A71b, A71d Small dot nozzle rows A100 Recording medium

Claims (10)

An image is recorded using a recording head having a large dot nozzle array configured by arranging nozzles for recording large dots on a recording medium and a small dot nozzle array configured by arranging nozzles for recording small dots. In the image recording apparatus to
Recording characteristic acquisition means for acquiring, for each area, recording characteristic information relating to the recording characteristics of the large dot nozzle array and the small dot nozzle array that perform dot recording in the same area of the recording medium;
A distribution rate determining means for determining a distribution rate of large dots and small dots for each of the areas based on the recording characteristic information;
A distribution pattern storage means for storing a plurality of distribution patterns defining pixels for recording large dots and small dots in the region in association with a combination of a quantized value of image data and the distribution ratio;
A distribution pattern setting means for selecting one of the plurality of distribution patterns from the quantization value and the distribution ratio, and setting the distribution pattern for each region;
Recording means for recording dots on a recording medium using the large dot nozzle row and the small dot nozzle row according to the distribution pattern set for each region by the distribution pattern setting means,
Among the plurality of distribution patterns, in the plurality of distribution patterns corresponding to the same quantization value, the arrangement of pixels for recording either large dots or small dots is constant regardless of the distribution ratio. A characteristic image recording apparatus. The recording head has a plurality of the large dot nozzle rows and a plurality of the small dot nozzle rows,
The distribution pattern storage means stores a distribution pattern that defines pixels to be recorded by each of the plurality of large dot nozzle rows and the plurality of small dot nozzle rows in the region,
The distribution pattern setting means sets the distribution pattern for each of the plurality of large dot nozzle rows and the plurality of small dot nozzle rows based on the quantization value and the distribution ratio. Item 8. The image recording apparatus according to Item 1. In the plurality of distribution patterns having the same distribution ratio and corresponding to different quantization values,
Pixels defined to record large dots with the distribution pattern corresponding to lower quantization values are defined to record large dots with the distribution pattern corresponding to higher quantization values,
The pixel defined to record small dots with a distribution pattern corresponding to a lower quantization value is defined to record small dots with the distribution pattern corresponding to a higher quantization value. 3. The image recording apparatus according to 1 or 2.   The image recording apparatus according to claim 1, wherein the plurality of distribution patterns are created so that large dots are arranged at a constant spatial frequency.   5. The image recording apparatus according to claim 1, wherein the recording characteristic information is information on an ejection amount of the large dot nozzle row and the small dot nozzle row.   5. The image recording apparatus according to claim 1, wherein the recording characteristic information is information on a diameter of an ejection port of the large dot nozzle row and the small dot nozzle row.   5. The image recording apparatus according to claim 1, wherein the recording characteristic information is information acquired by an image reading unit reading a test pattern recorded by the recording head. 6.   The recording head is configured by arranging a plurality of chips on which the large dot nozzle array and the small dot nozzle array are arranged, and the area is an area of a recording medium on which each of the chips records. The image recording apparatus according to claim 1, wherein the image recording apparatus is an image recording apparatus.   The image recording apparatus according to claim 1, wherein the image recording apparatus is an inkjet line printer. An image is recorded using a recording head having a large dot nozzle array configured by arranging nozzles for recording large dots on a recording medium and a small dot nozzle array configured by arranging nozzles for recording small dots. In the image recording method to
A recording characteristic acquisition step of acquiring, for each area, recording characteristic information regarding the recording characteristics of the large dot nozzle array and the small dot nozzle array that perform dot recording in the same area of the recording medium;
A distribution rate determining step for determining a distribution rate of large dots and small dots for each of the areas based on the recording characteristic information;
A distribution pattern storing step for storing a plurality of distribution patterns defining pixels for recording large dots and small dots in the region in association with a combination of a quantized value of image data and the distribution ratio;
A distribution pattern setting step of selecting one of the plurality of distribution patterns from the quantization value and the distribution ratio, and setting the distribution pattern for each region;
A recording step of recording dots on a recording medium using the large dot nozzle row and the small dot nozzle row according to the distribution pattern set for each region by the distribution pattern setting step;
Among the plurality of distribution patterns, in the plurality of distribution patterns corresponding to the same quantization value, the arrangement of pixels for recording either large dots or small dots is constant regardless of the distribution ratio. A characteristic image recording method.

Images