JPH06336016A - Ink jet recording method and recorder - Google Patents

Abstract

PURPOSE:To obtain ink jet recording method and recorder realizing the improvement of a bad effect on an image, unevenness of color, and the like caused by a positional shift of dots. CONSTITUTION:In an ink jet recorder, a multihead 702 including four color heads 701 is provided on a carriage, and a printing image is being completed on printing paper 707 by conducting a plurality of times of reciprocative recording operations on the same image area while paper is relatively and successively fed by a paper feed roller 703 and an auxiliary roller 704. Divided thinned-out pixel arrangement patterns used for every recording operation are compensated with each other. A division printing means is provided for making the divided thinned-out pixel arrangement pattern by arranging basic pixel groups, each of which is a rectangle (m) pixels long by (n) pixels wide meeting m<n, in accordance with a predetermined arranging rule.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an ink jet recording method and recording apparatus.

[0002]

2. Description of the Related Art With the widespread use of copying machines, word processors, information processing equipment such as computers, and communication equipment, an ink jet type recording head is used as one of image forming (recording) equipment of these equipment. Those that record digital images are rapidly spreading. In such a recording apparatus, in order to improve the recording speed, a recording head in which a plurality of recording elements are arranged in an array (hereinafter referred to as a multi-head) is formed by integrating a plurality of ink ejection ports and liquid paths. It is generally used and further provided with a plurality of the above-mentioned multi-heads for color correspondence.

FIG. 7 shows the construction of the printer section when printing on the paper surface by the above-mentioned multi-head. In this figure, reference numeral 701 is an ink cartridge. These are ink tanks filled with four color inks, black, cyan, magenta, and yellow, respectively,
702 multi-head. FIG. 8 shows a state of the multi-nozzles arranged on the multi-head from the z direction, and 81 is a multi-nozzle arranged on the multi-head 702. In this figure, the multi-nozzle 8
01 are arranged in parallel along the Y-axis, but they may have some inclination on the XY plane in the figure, for example. In this case, while the head moves in the traveling direction X, each nozzle performs printing while shifting the timing. Return to FIG. 7 again. A paper feed roller 703 rotates in the direction of the arrow in the figure while holding the print paper 707 together with the auxiliary roller of 704, and feeds the print paper 707 in the y direction at any time. Further, reference numeral 705 denotes a paper feed roller which feeds the printing paper and also plays a role of suppressing the printing paper 707 similarly to 703 and 704. Reference numeral 706 is a carriage that supports four ink cartridges and moves them with printing. This is when not printing
Alternatively, when performing a multi-head recovery operation or the like, the home position (h) at the position shown by the dotted line in the figure is waited.

Before the start of printing, the carriage (706) at the position shown in the figure (home position) moves in the x direction when the print start command arrives, and moves along the multi-head (702).
The upper n multi-nozzles (81) provide a width D on the paper.
Just print. When the data printing is completed up to the edge of the paper, the carriage returns to the original home position and x
Print in the direction. Or if it is double-sided printing, −
Printing is performed while moving in the x direction. From the end of the first printing to the start of the second printing, the paper feed roller (703) rotates in the arrow direction to feed the paper in the y direction by the width D. In this way, printing of only the multi-head width D and paper feeding are repeated for each scan of the carriage, whereby data printing on one paper surface is completed.

However, unlike a printer that prints only characters as a monochrome printer, various elements such as color developability, gradation, and uniformity are required to print a color image image. Particularly regarding uniformity, a slight variation in nozzle unit caused by a difference in multi-head manufacturing process affects the ink ejection amount and ejection direction of each nozzle when printing, and finally the printed image. This causes deterioration in image quality as uneven density.

A specific example will be described with reference to FIGS. In FIG. 9-a, reference numeral 91 is a multi-head, which is similar to that of FIG. 8, but for the sake of simplicity, it is assumed that it is composed of eight multi-nozzles (92). Reference numeral 93 is an ink droplet ejected from the multi-nozzle 92. Normally, it is ideal that ink is ejected in a uniform direction with a uniform discharge amount as shown in this figure. If such ejection is performed, dots of uniform size are landed on the paper surface as shown in FIG. 9-b, and a uniform image without density unevenness is obtained as a whole. (9-c). However, in reality, as described above, each nozzle has a variation, and if printing is performed in the same manner as above, as shown in FIG. The ejected ink drops vary in size and direction, and land on the paper surface as shown by 10-b. According to this figure
Area factor 1 periodically with respect to the head main scanning direction
There is a blank portion that cannot be filled with 00%, or conversely dots are overlapped more than necessary, or a white streak appears in the center of the figure. The collection of dots landed in such a state has the density distribution shown in FIG. 10-c in the nozzle arrangement direction, and as a result,
Normally, these phenomena are perceived as density unevenness as seen by human eyes.

Therefore, the following method has been devised as a countermeasure against this density unevenness. The method will be described with reference to FIGS. 11 and 12. According to this method, the multi-head 91 is scanned three times to complete the print area shown in FIGS. 9 and 10, but the half area of the 4-pixel unit is completed in two passes. In this case, the 8 nozzles of the multi-head are divided into a group of 4 upper nozzles and 4 lower nozzles,
The dots printed by one nozzle in one scan are defined image data thinned to about half according to a predetermined image data array. Then, at the time of the second scan, dots are embedded in the remaining half of the image data to complete the printing of the 4-pixel unit area. The recording method as described above is hereinafter referred to as a divided recording method. If such a divided recording method is carried out, FIG.
Even if the same recording head as that used in the above is used, the influence on the printed image peculiar to each nozzle is reduced by half, so the printed image looks like 11-b and the black as seen in 10-b. The streaks and white streaks become less noticeable. Therefore, density unevenness is 11
As shown in -c, it is considerably relaxed compared with the case of FIG.

When performing such recording, the first scan and the second scan
In the scan, the image data is divided according to a certain arrangement so as to fill each other. In the past, this image data arrangement (thinning pattern) is as shown in FIG. It was most common to use something like. Therefore, in the unit printing area (here, in units of 4 pixels), the printing is completed by the first scan for printing the zigzag lattice and the second scan for printing the inverse zigzag lattice. 12- of FIG.
a, 12-b and 12-c show how the recording of a certain area is completed when using this zigzag and reverse zigzag pattern, respectively, as in FIGS. It was explained using. First, in the first scan, the staggered pattern ◯ is recorded using the lower 4 nozzles (12-a). Next, in the second scan, the paper is fed by 4 pixels (1/2 of the head length) and the reverse zigzag pattern is printed (12-b). In the third scan, 4 again
The paper is fed by only pixels (1/2 of the head length), and the staggered pattern is recorded again. (12-c) In this manner, the paper feed in units of 4 pixels and the recording in the staggered and reverse zigzag patterns are alternately performed in this manner, whereby the recording region in units of 4 pixels is completed for each scan. As described above, by completing printing with two different types of nozzles in the same area, it is possible to obtain a high-quality image without density unevenness.

Such a recording method has already been disclosed in JP-A-60-1.
It is disclosed in 07975 and USP4967203, and it is stated that its effect on density unevenness and connecting lines is also effective. Regarding the former, "the paper feed of each main scan is made smaller than the width of the main scan so that two adjacent
And a means for forming overlapping portions by overlapping the main scanning of one time and a means for arranging the print dots of the overlapping portions so as not to overlap in the two main scannings. " . According to the present case, as described above, the thinning mask may be set to "print the odd-numbered steps and the even-numbered steps alternately in every other column", but in addition, the odd-numbered steps and the second-time steps may be performed in the first main scanning. There are cases in which even-numbered stages are printed in the above scanning, or printing is performed randomly in each scanning, and the thinning mask and the paper feed width are not completely limited.

On the other hand, in the latter USP4967203, "a) in the first pass, only the upper half of the first band is printed with alternating pixel positions which are not adjacent in the horizontal and vertical directions, and b) the second pass. At a pixel that was not printed in the first pass in the first swath and in the lower half of the first swath in alternating horizontal and vertical non-adjacent pixels, and c) in the third pass. 1, print the pixels in the first swath that were not printed in the second pass,
We will make a first pass to the belt that follows immediately after that. " As described above, in this case, the alternating pixel array which is not adjacent in the vertical and horizontal directions is limited as the thinning mask for performing the divided recording.

As a configuration to be added in this case,
A recording method is disclosed in which several pixels are collectively formed as a pseudo pixel (super pixel) for gradation expression and multi-color expression, and alternating thinning printing that is not adjacent in the horizontal and vertical directions in units of pseudo pixel (super pixel) is performed. ing. According to this method, "Once the system for implementing the above method is installed in either the program software or the printer software, the program can be called with the combination of color numbers specified for superpixels. This print quality is achieved without unnecessarily complicating the task of creating a computer program to produce multiple colors. "And simplifies programming for multicolored presentations. As It also states that color bleeding within a superpixel is harmless because each superpixel is intended to be perceived as a single, homogeneous color.

By the way, in the above-mentioned divided recording, 1
The time cost of printing on the paper surface is high, and there is an adverse effect that the throughput must be reduced. In such a case, a method of reciprocating print scanning of the carriage can be considered in order to perform printing in a shorter time. According to this
Since the carriage scan which has been returned to the home position without performing any printing after the conventional one-scan printing is performed, all of the carriage scanning can be practically halved. In fact, there are not a few monochromatic printing methods that perform the above-mentioned reciprocal printing. However, the color inkjet device having the structure as shown in the present invention has not been realized yet due to the following factors.

The landing state of the recording ink which is generally used at present and the medium (paper) for printing the same is shown in the sectional view of FIG. Here, the case where two different color inks (dots) are absorbed (recorded) at almost adjacent positions with a time difference is shown. The point to be noted is that in the two-dot overlapping portion, the dots printed later tend to sink in the depth direction of the paper than the dots printed earlier. This is a stage where the dyes such as dyes in the ejected ink are physically and chemically bonded to the recording medium. Since the bond between the recording medium and the dyes is finite, there is a large difference in the bonding force depending on the kind of the dyes. As long as there is no ink, the amount of ink dye ejected earlier remains on the surface of the recording medium because it is prioritized, and the ink dye ejected later does not easily bond on the surface of the recording medium, and It is thought that it sinks in the direction and is dyed.

In this case, even if two types of ink are printed at the same landing point, the priority colors differ depending on the order of the two types of ink ejection, and as a result, two different colors are expressed with respect to human visual characteristics. It will be. In FIG. 7, reference numeral 701 is a four-color head at the print start position. The respective colors are arranged from the right in the order of black, cyan, magenta, and yellow, and in the forward scan, the color moves from here to the right direction indicated by the X axis, and simultaneously records. At this time, since the recording order on the paper surface conforms to the above arrangement order, for example, when a green (cyan + yellow) signal is input to a certain fixed area, ink is absorbed in each pixel in the order of cyan and yellow. It
Therefore, as described above, in this scanning, the cyan absorbed first becomes the priority color, and the green dot having a strong cyan tint is formed. On the contrary, in the backward scan after the paper is fed in the Y-axis direction, the four-color head is located on the right side of the drawing, and this time, recording is performed while moving in the direction opposite to the forward path. Therefore, the driving order is also reversed, and a green dot having a strong yellow tint is obtained by this scanning. By repeating the above scanning, green dots having a strong cyan tint and green dots having a strong yellow tint are printed according to the forward and backward passes of each print head. If each scan does not use the division printing method and the paper is fed by the head width for each reciprocating scan, the green area with a strong cyan tint and the green area with a strong yellow tint are the heads. The green image, which is alternately repeated widthwise, will be significantly deteriorated in the green image which should be uniform.

However, this problem can be overcome to some extent by using the divided recording method described above. In division printing, green dots having a strong cyan tint are recorded in the forward pass (12-a, c) and green dots having a strong yellow tint are recorded in the backward pass (12-b), as already described in FIG. . Therefore, the tint in a certain area is alleviated by the dots of both tints.

This structure and effect are described in USP4748453.
It has already been disclosed in. Although the paper feed amount is not limited here, it is possible to perform complementary printing on pixels that are alternately positioned in the horizontal and vertical directions in the same printing area by the first and second (or more) divided printing scans. , Tp, and the like, to prevent beading of ink on the medium, and when forming a color image, reverse the ink ejection order of the mixed color pixels between the first scan and the second scan (reciprocal recording) to achieve color banding. The effect that can prevent (color unevenness) is described. In this case, the main purpose is to prevent beading between pixels, so that the pixels printed in one scan are characterized by being alternating in the horizontal and vertical directions (not adjacent to each other). There is.

On the other hand, JP-A-58 of the same applicant as the present invention
In 194541, when a plurality of recording element arrays are arranged in parallel and reciprocally run in a direction orthogonal to the recording element arrays to perform dot matrix recording main scanning, each row of the recording dot matrix in the forward pass of the main scanning. And intermittently recording a smaller number of dots than at least one dot to be recorded in at least one of the columns, and intermittently recording the remaining dots in at least one of the rows and columns in the return pass of the main scanning. Therefore, the order of the overlapping of the recording in the overlapping recording dots by the plurality of recording element arrays is made different between the forward pass and the return pass of the main scanning ”. In this case as well, there is no limitation to reduce paper feed than usual as in the case of the divided recording described above, and the effect is to prevent image deterioration due to color tone deviation (color unevenness) of the recorded image due to overlapping recording of color inks. I am trying. In this case, since the main purpose is to prevent this color tone shift, there is no special limitation on the dot position printed in each scan, and in the embodiment, in addition to the checkered pattern (staggered, reverse zigzag), It describes horizontal thinning in which only alternate recording is performed and vertical thinning in which alternate recording is repeated only in the horizontal direction.

Although not limited to a color printer, Japanese Patent Application Laid-Open No. 55-113573 also discloses a structure in which reciprocal recording is performed by using a checkered pattern (staggered or inverted zigzag) pattern. In this case, it is an object to prevent adjacent dots from being continuously printed, thereby preventing the dot distortion by printing the adjacent dots before the printed dots are dried. Therefore, here is the above US
Similar to P4748453, the thinning mask is limited to a twill pattern.

By the way, all of the three patents listed here are aimed at preventing color unevenness and beading during reciprocal recording. Therefore, unlike the divided recording method described in the present case, the configuration of “making the paper feed amount between each scanning equal to or less than the normal head width” which is performed for the purpose of preventing density unevenness due to nozzle variation is not adopted. .

As described above, if the divided recording is performed in the reciprocal recording, two kinds of recording pixels whose ink colors are ejected in the opposite order can be evenly provided in the recording area, which causes color unevenness. It was possible to easily perform multi-color bidirectional recording.

[0021]

However, even if the division recording by the zigzag / inverse zigzag shown in the above-mentioned proposal is performed, the problem of color unevenness is not completely solved and it has not been put to practical use.
The reason for this will be described with reference to FIGS. Normally, the amount of ink droplets is designed to spread larger than the area given to each pixel on the paper surface. This is because the blank area cannot be seen at all in the area of 100% printing rate data. Therefore, even when the divided recording method is performed, only 50% of the recording pixels are printed, but the print medium (recording paper) is almost 100% as shown in FIG.
Area is covered. FIG. 20 shows a cross-sectional view of the plane of this case. Here, a case is shown in which staggered printing is performed on a blank sheet in the first pass (forward scan) and reverse zigzag printing is performed in the second pass (return scan). Reference numeral 2001 shows the state of the ink immediately after the first pass (forward pass) printing, in which the black-painted portion is cyan ink and the shaded portion is yellow ink. Since the yellow ink is ejected at the same position as the cyan ink with a slight time difference, when absorbed by the paper, the cyan ink is in a state with less bleeding and a higher density, and the yellow ink is below the cyan ink. It bleeds so much that it wraps around the periphery, resulting in a print with a low density. At this time, these inks are absorbed up to the adjacent pixels, and the paper surface is almost completely filled with the ink as shown in FIG.

Second pass (return pass) performed under these conditions
As shown in 2003, the ink is landed on the adjacent ink which has already been absorbed. Since the second pass is the backward scan, yellow is printed first and cyan is printed later (200
2). If the ink is absorbed as it is, it will eventually reach 20
Both colors such as 03 are in an absorption state that does not appear much on the surface. Then, the density of cyan that is printed first is most strongly emphasized in the final print-completed image, and this print area becomes a green image in which the tint has priority to cyan. On the contrary, in the print area adjacent to the above-mentioned print area where the backward scan is the first pass, the positions of cyan and yellow are reversed, and a green image with a tint of yellow is obtained.

FIG. 6 shows how the above two print areas appear. As can be seen from this figure, the lower half of the head always determines the priority color of each area, and the priority color is reversed by reciprocating scanning. Since these two areas having different priority colors are alternately present, color unevenness still appears even in division printing, deteriorating the image, and reciprocal printing is practically difficult.

Further, the adverse effect of ink bleeding on adjacent pixels has been confirmed not only in the above-mentioned color unevenness but also in single-color bidirectional printing. This will be explained below. FIG. 21 shows FIG.
Similar to 0, the ink absorption states of the first and second passes are shown for a single color. In this figure, 2101 represents the landing state after the first pass printing, 2102 and 2103.
Both show the state after the second pass printing in a cross-sectional view of the paper. Here, 2102 is 2 immediately after the first pass is recorded.
A state in which the second pass is recorded, and 2103 is a state in which the second pass is recorded after a certain time has elapsed after the first pass recording. A difference appears in the state of absorption of the ink recorded in the second pass on the paper surface between them. 2102 is absorbed in the depth direction of the paper surface, whereas in 2103, the ink of the second pass spreads on the paper surface. This is also confirmed from the back side of the paper, and 2103 is 2
The degree of ink leakage to the back side of the paper is larger than that of 102. Then, such a state appears as a density difference on the paper surface between the two (2104, 2105).

The time difference caused by the reciprocating operation of the carriage is sufficient compared to the order of the time difference for generating the density difference between the two. Then, this factor appears as a new adverse effect due to the reciprocal printing. The reason will be described below with reference to FIG.

In FIG. 22, the head first scans forward from the position 2201 in the direction of the arrow to perform recording with the first scan width. Then, when printing for one line is performed, the paper is fed by ½ of the scan width, and backward scanning is performed from the position 2202 in the drawing in the reverse direction. Further, after feeding the paper having the same width as the above, recording is performed in the direction of the arrow again from the position 2203 as forward scanning. The following is a comparison of the recording intervals of the two passes for each of the positions 1 to 3 of the printing area completed at this time. That is, in and, the printing of the second pass is performed immediately after the 1/2 width paper feed is performed after the printing of the first pass is completed. On the other hand, and in 1
After the pass recording, the second pass is recorded after the time for the carriage to make one reciprocating scan. And, the part recorded at the time difference which is exactly in the middle is.
Therefore, as described above with reference to FIG. 21, and are the areas of highest density among these, and are followed by, and and are areas of low surface density where ink has been deeply absorbed. Therefore, density unevenness appears in the left-side area where ½ widths are repeated in the vertical direction and the right-side area where ½ widths are repeated, which causes an adverse effect on the image.

As described above, the bleeding to the non-printed pixels in the first pass causes the density to be affected by the recording interval in the first pass and the second pass. It is impossible and I can understand that I was fit. Although the above description has been made by taking the single-color recording as an example, this phenomenon appears with the color unevenness already described in the case of the mixed-color recording. Recognized as a difference. Also in the one-way recording, the following are harmful factors that affect the recording time difference. If the printing apparatus performs head recovery scanning to maintain its own driving scanning during printing, or if it is waiting for the print data to be transferred, the carriage temporarily stops. Then, such a pause produces irregular density unevenness in an order of magnitude larger than the time difference unevenness described above. That is, the carriage enters a rest state while the first pass is being printed, and after a lapse of time, the printing area printed has a darker density than the other areas. The density unevenness caused by such factors is distinguished from the time difference unevenness and is hereinafter referred to as rest unevenness.

Further, the present inventors have confirmed that there are different problems in addition to the above problems. The problem will be described below with reference to the drawings.

FIG. 23-a shows a state in which the head is moving at a constant speed V in the forward or backward direction and ejecting an ink drop at a constant speed v onto a smooth paper surface. If the paper surface is smooth as shown and the distance d between the paper surface and the head face surface is kept constant at a predetermined value,
By uniquely adjusting the ejection timings for the forward and return passes, the dots printed on the forward pass and the dots printed on the return pass are landed at the same position on the paper as shown in the figure below. However, if the paper surface itself is lifted from the actual position like 23-b due to some reason, the distance between the head face surface and the paper surface is reduced to d ′, and the ink drops arrive at the paper surface after the head ejects. Since the time until is shortened in both the forward and backward passes, the print dots are landed at different positions, which are different from the target positions, as shown in the figure below. Similarly, when the distance d between the paper surface and the head face surface at the time of forward printing for the same image area and the distance d at the time of backward printing change due to local floating of the paper, a larger landing position shift occurs. May occur.

In this state, when the 100% duty image is printed in both directions using the thinning mask described above, the dot landing state becomes as shown in FIG. In addition, FIG. 26 shows a case where divided recording is printed in both directions using a conventional staggered mask. Here, each dot is shown shifted by 1/4 pixel from the normal position, and the portion where adjacent dots overlap more than necessary and the portion where a gap is formed between adjacent dots differ depending on the thinning mask. It appears in the arrangement. In FIG. 26, since all the dots are printed in the opposite direction to the adjacent dots, a gap of one dot is generated for each dot, and the density is generally low.

Such a landing position shift during bidirectional printing is caused not only by partial ups and downs of the paper surface shown in FIG. 23, but also by various causes such as uneven ejection speed of the recording head and unevenness of carriage movement speed. Is. Since the values of these factors are not always constant in the traveling direction of the carriage, it is difficult to control the ejection timing during bidirectional printing. Further, the distance between the head and the paper surface of the recording apparatus may vary widely from one apparatus to another, so there is a limit to the reciprocating landing position control by adjusting the ejection timing.

The present invention has been made in order to solve the above-mentioned problems of the prior art, and at the same time, improves the image adverse effects due to the dot positional deviation during bidirectional printing and the image adverse effects such as color unevenness, pause unevenness, and time difference unevenness. An object of the present invention is to provide an ink jet recording method and a recording apparatus that can be used.

[0033]

Therefore, an ink jet recording method and a recording apparatus according to the present invention use a multi-head in which a plurality of ink ejection ports are arranged, and perform reciprocal recording scanning a plurality of times for the same image area. In the ink jet recording apparatus that completes a print image on the paper surface by performing relative sheet feeding, the divided thinned pixel array in each recording scan has a complementary relationship, and the divided thinned pixel is The arrangement is intended to achieve the above-mentioned object by a constitution characterized in that it has a division print control means in which rectangular basic pixel groups of m <n consisting of vertical m pixels and horizontal n pixels are arranged in accordance with a predetermined arrangement rule. It is a thing.

[0034]

According to the present invention, a multi-head in which a plurality of ink ejection ports are arranged is used, and reciprocal recording and scanning are performed a plurality of times for the same image area, and the paper is relatively sequentially fed so that the paper surface In an ink jet recording apparatus that completes a printed image on top, the divided and thinned pixel arrays in each print scan have a complementary relationship, and the divided and thinned pixel arrays are m <n composed of vertical m pixels and horizontal n pixels. Inkjet recording apparatus comprising divided print control means in which rectangular basic pixel groups of 4 are arranged according to a predetermined arrangement rule, and more preferably, an arrangement of rectangular basic pixel groups of m <n consisting of the vertical m pixels and horizontal n pixels. According to the rule, the number p of pixels continuously arranged in the vertical direction is at least m.
By using an inkjet recording device having an arrangement rule satisfying <p <n or a multi-head having a plurality of ink ejection ports arranged, a plurality of reciprocal recording scans are performed on the same image area and relatively sequentially. In an inkjet recording method for completing a printed image on a paper surface by performing paper feeding, the divided and thinned pixel arrays in each recording scan have a complementary relationship, and the divided and thinned pixel arrays are vertical m pixels, An ink jet recording method characterized by arranging rectangular basic pixel groups of m <n consisting of horizontal n pixels according to a predetermined arrangement rule, more preferably, a rectangle of m <n consisting of vertical m pixels and horizontal n pixels. In the arrangement rule of the basic pixel group, the arrangement rule is such that the number p of pixels continuously arranged in the vertical direction satisfies at least m <p <n. By providing the inkjet recording method, it is possible to provide an inkjet recording method and a recording apparatus capable of simultaneously improving the image adverse effects due to the dot position deviation during bidirectional printing and the image adverse effects such as color unevenness, pause unevenness, and time difference unevenness. .

[0035]

EXAMPLE A first example will be described below. FIG. 27 shows the dot landing state of this embodiment, and here, 1 × 4.
Rectangles are used as basic pixel groups, and they are arranged in a staggered manner so as not to be adjacent to each other in the same scan.

The object of the present invention is to prevent adverse effects of bidirectional dot position deviation which is caused by ups and downs of the paper surface, various driving irregularities, and changes in ejection speed without deteriorating the color irregularity which has already been improved in bidirectional printing. Is to make it as inconspicuous as possible. Therefore,
In general, what kind of thinning-out mask arrangement can be used to achieve these will be described below.

Misalignment during bidirectional printing always occurs under the same condition for any dot. Therefore, how to reduce the gap as shown in FIG. 26 is one of the factors that determine the objective effect of the present invention.

The number of gaps is equal to the number of spots where the dots landed on the outward path and the dots landed on the return path are adjacent to each other.
In FIG. 26, the forward printing and the backward printing are alternately arranged every other dot, so that a gap is formed next to all the dots. However, when a group of four consecutive dots in the lateral direction is printed at the same time as shown in FIG. 27 of the present embodiment, no gap is formed in these four dots, and the number of gaps is set to once every four. It can be suppressed.
That is, the number of gaps of dot misalignment during bidirectional printing is related to the horizontal length of the basic dot group.

FIG. 30 shows how dot misalignment affects an image when two-way binarization method (dither method) is used for reciprocal printing using the thinning arrangement of this embodiment. It was done. Up to now, it has been 1 as a measure of image quality.
Although the way of opening the gap with a duty of 00% has been mainly increased, in fact, the most noticeable dot misalignment in both directions is about 50% halftone. In this halftone, the density and granularity of the image or the appearance of image deterioration due to dot misalignment during bidirectional printing will be somewhat different depending on various dither methods. Figure 3 now
0 indicates a case where the dots in bidirectional printing are shifted by 1/2 pixel and a dot is shifted by 1 pixel for two types of dither A and dither B. As can be seen from this figure, the dots are displaced by the same amount. However, it can be seen that the vertical lines are conspicuous in the dither A, while the dither B maintains a relatively uniform state. Therefore, when the 1 × 4 thinning-out array of this embodiment is used, the type such as dither B can keep the image relatively good. FIG. 31 shows the two dither methods shown in FIG. 30 as threshold values within 8 × 8 pixels. Although not shown here, it is confirmed in the present embodiment that the image using the dither B is more stable and the image is stable at any duty than the case using the dither A. .

As described above, in this embodiment, 1
By arranging the basic pixel groups of × 4 so as not to be adjacent to each other, the influence of dot deviation during bidirectional printing can be reduced and a uniform and smooth image can be obtained.

Although the ink color or the number of ink colors is not particularly limited in this embodiment, the effect of this embodiment is effective in all cases.

Further, in this embodiment, four pixels are continuous in the horizontal direction, but only one pixel is held in the vertical direction. Therefore, with respect to the color unevenness, the pause unevenness, and the like described in the conventional example, when the dot diameter is large, the bleeding into the non-printing area is large and the effect is difficult to appear. It can be said that a recording medium such as an OHP sheet having a small absorption, that is, less color unevenness and a small dot diameter is suitable for the configuration of the present embodiment.

(Other Embodiments) Next, the second embodiment will be described. The present embodiment is characterized in that, as the thinning-out mask, rectangular pixel groups whose vertical direction is larger than that in the first embodiment are arranged so as not to be adjacent to each other. When such a mask is used, the amount of ink bleeding out to the non-printing area in one scan is reduced, and the unevenness in time difference and pause due to the time difference between each scan and the recording order of color ink are also caused. It is possible to reduce the color unevenness.

This embodiment will be specifically described below with reference to the drawings. FIG. 2 better illustrates the features and effects of the method by comparing with FIG.
The ink landing state after scanning is shown. In the conventional example, the dots printed by the same scan are landed on the alternating pixels which are not adjacent to each other (FIG. 3), but in this example, all the pixels are 3 ×.
The pixel groups are arranged in such a manner that they are recorded in the same pass and the groups are recorded in positions which are not adjacent to each other in the same pass. By doing so, as is clear from the figure, the ink overlapping area of the first recording color is reduced by the increase of the overlapping area of the dots in each group, so that the white paper part that can be landed after the second recording color is reduced. To increase. Therefore, even if the ink ejection order is changed by the reciprocal scanning, the area ratio occupied by the priority colors of the first scan is reduced, and the area occupied by the priority colors of the second scan is increased accordingly, so that there is a deviation between the priority colors. This eliminates the difference in the tint of the area that is repeated for each paper feed width.

Below, FIG. 2 is used as a print explanatory diagram of this method.
The recording method using the dot group of 3 × 4 pixels shown in FIG. FIG. 5 is a recording state diagram in this patent compared with the conventional example of FIG. With a head composed of 16 nozzles, bidirectional printing is performed while feeding paper by 8 nozzles. The data recorded here is a 100% green image composed of cyan and yellow. In these figures, the case where dots having a diameter of 110 μm are recorded for a pixel density of 360 dpi is shown as an example.

In FIG. 6, since the print mask has a zigzag pattern in each scan, each dot having a size of one pixel or more bleeds into an adjacent pixel, and density unevenness having a priority color different for each feed width is obtained. Has occurred. On the other hand, in FIG. 5 of the present method example, since the 3 × 4 staggered mask is used, the amount of ink bleeding in each dot group,
The bleeding into other areas is limited, and the difference in priority color in each area is considerably suppressed as compared with FIG.

By the way, in applying the method of this embodiment, the size of the mask, that is, the appropriate value of m × n varies depending on the given pixel density, ink ejection amount, conditions of the recording medium and the like. If m or n is too small under the above conditions, the effect of this method cannot be obtained, whereas if the dot group is too large to be resolved by the human eye, the color tint between adjacent groups will be increased. Difference is detected, and the image becomes rough and rough.

The present invention is also effective for unevenness in time difference or unevenness in rest, which is confirmed even in a single color. As already described in the conventional example, the time difference unevenness is found when the second pass is printed in the area where the dots of the first pass are protruding and bleeding. Therefore, as in this method, if the dots in the first pass are adjacent to each other and the protrusion ratio is small, it is possible to print in two-pass printing under almost the same conditions as in one-pass printing. The factors that are caused will disappear. FIG. 15 shows FIG.
The above effect is shown by emphasizing the overlapping portion of the first scan and the second scan. It can be seen that the total area amount of the overlapping portion with respect to the recording area is obviously smaller in this embodiment than in the conventional example.

By the way, basically, the divided printing method does not exhibit its effect unless the printing pixels in the unit area are printed in substantially equal portions in two scans. In the example I've explained so far,
Since 100% duty images were all recorded, an equal number of pixels were always recorded in two passes in both FIG. 5 and FIG. However, the image data that is actually sent as a signal is multi-valued data that represents a certain gradation.
Many of them are binarized by a predetermined binarization method, and are sent after being set to a predetermined pattern. The present method is effective for the dither method which is used most often among them. One example thereof is shown in FIG. Among the dither methods, the Bayer type is particularly used as the binarization method. When a duty image in 1/16 (4/64) increments is given in a fixed area of 8 × 8, each divided image data is compared with the previous divided recording and the divided recording method performed in this method example. It shows how it is divided into two paths. The print pixels allowed in the first scan and the second scan by the divided printing method of this embodiment and the conventional divided printing method are shown in black. Then, when the binarized image data shown on the left side is input, which pixel is printed by each scan of each printing method is shown on the right side. According to this, it can be seen that in the conventional division printing, the printing is completed by only one scanning up to 8/16, that is, 50% duty. Also, 5
Even if it exceeds 0%, the deviation in the number of print pixels in each pass is large, and the number of both becomes equal only when 100% is reached. When such a situation occurs, for example, at a low duty of 50% or less, data of one pixel width in the main scanning direction of the head is all printed by the same nozzle, which is the original purpose of the divided printing method described above. The density unevenness due to the nozzle variation is not completely eliminated. Furthermore, the non-uniformity of the print pixels in the first scan and the second scan is 50%.
It is expected that even in the high duty region exceeding (8/16), the image is considerably inferior in the density unevenness as compared with the case of the present embodiment in which the duty is always equalized. This problem is not limited to bidirectional printing, but can occur in unidirectional printing.
Further, as the same factor, that is, a problem phenomenon caused by unevenness of print pixels in each scan, when two-color inks having different duties other than the above-mentioned density unevenness are used for printing, the ink ejection already described in the conventional example is performed. It also causes color unevenness due to the order. Similar to the density unevenness due to nozzle variation, this problem also occurs in one-way printing, and can be solved by making it possible to equally distribute the number of pixels printed in each pass. 3 of this embodiment
If a thinning mask consisting of x4 dot groups is used, the number of dots between both scans is always divided equally.
Therefore, the color unevenness does not occur. Further, since the dots lined up in the same main scanning direction are always printed by two scans, that is, two kinds of nozzles, this method is effective even for density unevenness due to nozzle variations.

As described above, by using a 3 × 4 dot group as a thinning mask, color unevenness
It is possible to obtain a smooth and high-quality image without time difference unevenness, pause unevenness, and nozzle unevenness. The shape of the mask may be a square mask such as 3 × 4 described above, or a mask longer in the lateral direction may be used to prevent image deterioration due to dot misalignment during bidirectional printing.

As described above, in the color ink jet recording apparatus, the bidirectional printing and the unidirectional printing in the color ink jet recording apparatus can cause color unevenness and time difference due to the previously proposed divisional recording method and the disclosed patent. It is possible to perform printing without any adverse effects such as unevenness, unevenness in rest, and unevenness in nozzles.

In this embodiment, the basic pixel group is about three times larger than that in the first embodiment, so the effect on color unevenness will be considerably large. However, if only the color unevenness is emphasized too much and a too large basic pixel group is applied, a lump of each basic pixel group itself becomes conspicuous and the image may be rough. Also, in this embodiment, as in the first embodiment, four consecutive pixels in the horizontal direction are printed by the same printing scan, so the number of gaps caused by dot deviation during bidirectional printing is the same as in the first embodiment. Yes (FIG. 25), the number of gaps is reduced to 1/4 compared to the conventional example (FIG. 26). However, in this case, since the gaps are continuous in the vertical direction by three pixels, it may be conspicuous as a texture depending on the size of the gap.

Depending on various factors as described above, the number of pixels in the horizontal direction, the number of pixels in the vertical direction, and the total number of pixels in the mask,
Further, as in the shape of the mask, the thinning mask for realizing the present invention can realize an appropriate pattern each time.

The third embodiment will be described below. FIG. 1 shows a dot landing state of a thinning-out mask of this embodiment, which is applied simultaneously with four colors, in comparison with FIGS. Further, FIG. 4 is similar to FIGS. 5 and 6 and shows the green 10 when the above mask is used.
The state of formation of a 0% image is shown.

In FIGS. 3 and 6, the conventional method in which the dots printed by the same scan are landed on the alternating pixels which are not adjacent to each other,
In FIGS. 2 and 5, all the pixels are grouped into a 3 × 4 group, the dot groups are printed in the same pass, and the groups are printed in positions that are not adjacent to each other in the same pass. Is the method. On the other hand, in FIGS. 1 and 4 of the present embodiment, 1 × 4 is set as a basic dot group, and this is recorded by a mask which is arranged adjacent to each other in the vertical direction while being shifted by 2 pixels or more in the horizontal direction.

2 and 5, the main purpose is to increase the overlapping area of the dots in each group to reduce the ink covering area of the first recording color as much as possible.
As described above, this method is mainly effective in preventing color unevenness during bidirectional printing. On the other hand, in FIG. 1, since the number of dots arranged in the vertical direction is as small as 2 dots or less, the dot overlapping area in each group is somewhat smaller than that in FIG. But,
As can be seen by comparing FIG. 4 with FIGS. 5 and 6, the first recording color and the second recording color of the green image on the paper surface are equally divided even in the present embodiment, and the color unevenness is hardly harmful. I won't.

FIG. 24 is a diagram best showing the effects of the present embodiment of the present invention. Also in this figure, as in FIGS. 25 and 26, in the bidirectional printing, the dot position deviation in the bidirectional directions described above has occurred by ¼ pixel. Compared to the other two figures, the number of gaps in this embodiment is small and the gaps are evenly distributed, so that the density becomes thin as shown in FIG. 26 and the vertical texture as shown in FIG. Not noticeable.

Next, how to arrange a fixed number of gaps inconspicuously will be described. In both FIG. 24 and FIG. 25, four dots are printed in the horizontal direction, so that the number of gaps in the entire image is the same. However, in FIG. 25 of the second embodiment.
Then, the gaps of 4 dots are continuously arranged in the vertical direction, and the place and size thereof are clearly confirmed.
On the other hand, in FIG. 24 of the present embodiment, since the 1 × 4 basic dot groups are arranged in the vertical direction while being shifted by 2 pixels or more in the horizontal direction, the gaps are not adjacent to each other, and the unit of pixel unit is 1 pixel. Small gaps are evenly distributed on the image. Such small gaps are not only inconspicuous, but often do not appear depending on the dot size and the amount of misalignment, and are evenly distributed, so that they do not cause any harmful effects as a whole image.

As described above, in order to prevent the gaps from being continuously arranged in the vertical direction with the basic dot groups fixed, the basic dot groups adjacent to each other in the vertical direction are always arranged at positions shifted by one pixel or more in the horizontal direction. There is a need. In the case of the present embodiment, since 1 × 4 basic dot groups are used, vertically adjacent basic dot groups must be displaced to the left or right within the range of at least 1 pixel and at most 4 pixels. Of.

As described above, the following two conditions are given as the conditions given to the arrangement method of the basic dot groups in order to make the dot misalignment in bidirectional printing inconspicuous.

(1) Increase the number of dots in the horizontal direction of the basic dot group.

(2) Basic dot groups that are vertically adjacent to each other are arranged so as to be offset by one pixel or more on either side.

However, the above two conditions are intended only for dot misalignment during bidirectional printing, and it is said that by satisfying only these conditions, nozzle unevenness, color unevenness, time difference unevenness, etc., which have been harmful effects on images, can be overcome. Not exclusively. For example, if the lateral direction of the basic dot group is too long in order to satisfy the condition (1), the nozzle unevenness, which is the original purpose of the divided recording method, cannot be alleviated. Also,
When 1 × 4 pixels are used as a basic dot group as in this embodiment and the left / right deviation amount of (2) is fixed to 4 pixels, the result is as shown in FIG. 27 of the first embodiment. In both cases, when the dots are large, the ink coverage of the first recording color is large, and color unevenness is likely to occur. Further, if the basic dot groups are arranged while always shifting in the same direction as shown in FIG. 28, there is no harmful effect due to color unevenness or dot misalignment, but as shown in the figure, a regular texture in a diagonal arrangement can be noticeable. There is a nature.
On the other hand, in the case of the mask patterns that are vertically adjacent to each other by shifting by 2 pixels or more and 4 pixels or less as described in the present embodiment, the ink coverage is low and the texture is not conspicuous while eliminating unevenness due to dot misalignment. Can be realized.

Further, FIG. 17 shows the pixel positions printed by two scans when divided printing is performed using this mask, as in FIG. In this figure as well, the numbers of dots in the first scan and the second scan are always equally divided.
With respect to the color unevenness caused by the deviation of the number of print dots between each scan, the same effect as that of the 4 × 4 mask described above can be obtained. Since the mask of 4 pixels in the horizontal direction is used, the dots arranged in the same main scanning direction are equally divided into two scans. Like the 4 × 4 mask, it is often effective in this respect not only for the Bayer type but also for other dither methods.

As described above, in this embodiment, m =
By arranging the basic dot groups m × n of 1 and n = 4 in a stepwise manner as shown in FIG. 1, the number p of pixels continuous in the vertical direction always satisfies m <p <n. This makes it possible to make the gaps that are explained inconspicuous.
In addition, the thinning arrangement as shown in FIGS. 1 and 4 is black,
By using it for both cyan, magenta, and yellow, and using the bidirectional split recording method, it is possible to obtain high-quality and smooth images that also eliminate the adverse effects such as color unevenness, time difference unevenness, pause unevenness, and nozzle unevenness. ing.

Further, in this embodiment, the basic dot group (m × n) has been described as 1 × 4 pixels, but it is not fixed to this value both vertically and horizontally. As described above with respect to the horizontal direction (n), as long as n is large, the number of gaps between dots decreases as long as the adverse effect of nozzle unevenness does not appear. In the vertical direction (m), m must be 1 in order to completely avoid the continuous arrangement of the gaps as shown in FIG. 25. However, depending on the image resolution, the ink ejection amount, and the color unevenness state, m = 1. May be 2 or more. However, generally, unless m <n, the effect of the present invention is hard to appear.

It is also possible to expand the present embodiment to make the thinning mask different for each color and each scanning as shown in FIG. Here, the same thinning masks as in the first embodiment are used for cyan, magenta, and yellow, but for cyan, masks complementary to the other two colors are alternately used in each scan. Black uses a mask that is completely different from other colors.

In many cases, black is always the priority color regardless of the order in which the colors are imprinted, while the dot position deviation in both directions is particularly noticeable at 100% duty. Further, in order to make the vertical ruled lines and the horizontal ruled lines smooth, it is preferable not to arrange large masks alternately in the vertical direction and the horizontal direction as shown in FIG. According to the mask shown in FIG. 13, the pixels arranged in the vertical direction have the forward print dots and the backward print dots landed alternately by one pixel, so that the vertical ruled line also has a straight line with no noticeable dot deviation in both directions. Further, in the horizontal direction, since printing is performed by almost one type of nozzle, there is no disturbance of the horizontal ruled line due to variations in the paper feed amount. In this case, the effect on not only the color unevenness but also the nozzle unevenness is reduced as compared with the other colors, but the linearity and the character quality are considerably improved.

In this case, in black, the nozzle unevenness is eliminated only at a ratio of 1/16 pixel in the main scanning direction.
Depending on the nozzle unevenness and the gap unevenness in both directions,
9 (Example 1), the number of gaps increases, but it may be set in a direction to eliminate nozzle unevenness. Further, as shown in the same drawing (Example 2), even if the number of gaps is the same as that of black in this embodiment, the positions of the gaps may be varied over the entire image.

This method attaches particular importance to color unevenness in color inks and actively attempts to solve this.
A fundamental cause of color unevenness is that the color heads are arranged in four colors in the same direction as the carriage scanning direction. Since such a configuration is adopted, there is inevitably a difference in the printing order in each scan between the respective colors. Therefore, the preceding example of the above-described third embodiment is one in which the recording order is alternately reversed and recording is performed under the same conditions as much as possible in the forward pass and the return pass in which dots of different tints are formed. On the other hand, in the present method, the aim is to prevent the difference in priority color between each color as much as possible even in each scanning, and in order to realize this, the combination of two colors in which color unevenness is easily noticeable is used. The mask is reversed. FIG. 14 shows a state in which the thinning mask of FIG. 13 is used and printing is performed while exchanging cyan and yellow for each scan. Also in the case of this method, it can be seen that there is no uneven color tone in each region, as in FIG. If different masks are used for each color like this, before the ink (first cyan for forward scan and yellow for backward scan) that is printed first on a blank sheet is completely absorbed by the paper surface, another Color inks (yellow for forward scan, cyan for backward scan) are printed, so the printing conditions for each color within each scan are made more uniform,
Differences in priority colors are less likely to appear.

Of course, in this method, since printing must be completed by two carriage scans for each area, it is impossible to use a mask in which all three colors do not overlap at all. However, the number of color combinations in which color unevenness is recognized as an adverse effect on an image is often about one set or two sets. In such a case, the two colors may be set to be printed by different masks. Further, in the case where color unevenness is conspicuous evenly in a combination of each color, a mask in which each color is shifted by several pixels may be used.

However, in this method, since the mask is set in the direction in which all pixel areas are printed in the first scan of each area, the time difference between the first scan and the second scan becomes a detrimental factor. The time difference unevenness tends to worsen.
Therefore, the first embodiment is more effective in the case where the time difference unevenness is considerably conspicuous, whereas the present method is effective in the case where the color unevenness is particularly remarkable in the specific two colors in the present method. Is.

[0073]

As described above, according to the present invention,
By using a multi-head with a plurality of ink ejection ports arranged, reciprocal recording and scanning of a plurality of fields for the same pixel area, and relatively sequential paper feeding, a printed image is completed on the paper surface. In the ink jet printing apparatus, the divided thinned pixel array in each printing scan has a complementary relationship, and the divided thinned pixel array has m vertical pixels and n horizontal pixels.
By providing the divided print control means in which the rectangular basic pixel groups of m <n made up of pixels are arranged in accordance with the predetermined arrangement rule, the image misalignment due to the dot position deviation during bidirectional printing, the color unevenness, the pause unevenness, and the time difference unevenness It has become possible to alleviate such image adverse effects at the same time.

[Brief description of drawings]

FIG. 1 is an example of a divided recording thinning-out array according to a third embodiment of the present invention.

FIG. 2 is an example of a divided recording thinning array according to a second embodiment of the present invention.

FIG. 3 is an example of a divided recording thinning array of a conventional example.

FIG. 4 is a recording state of the third embodiment.

FIG. 5 is a recording state of the second embodiment.

FIG. 6 is a recording state of a conventional example.

FIG. 7 is a diagram of a printing unit of an inkjet printer used in the present invention.

FIG. 8 is a diagram of a multi-head used in the present invention.

FIG. 9 is a diagram showing an ideal printing state of an inkjet printer.

FIG. 10 is a diagram showing a printing state of an inkjet printer having uneven density.

FIG. 11 is a diagram showing division printing used in the present invention.

FIG. 12 is a diagram showing division printing used in the present invention.

FIG. 13 is an example of a divided recording thinning array of the third embodiment.

FIG. 14 is a recording state of the third embodiment.

FIG. 15 is a diagram showing a region affected by a round-trip time difference in the second embodiment of the present invention.

FIG. 16 is a diagram showing a region affected by a round-trip time difference in the conventional example.

FIG. 17 is a diagram of correspondence to binarized data in the third embodiment of the present invention.

FIG. 18 is a correspondence diagram for binarized data in the conventional example and the second embodiment 2 of the present invention.

FIG. 19 is a sectional view showing an ink absorption state on the paper surface.

FIG. 20 is a cross-sectional view of a paper when color unevenness occurs

FIG. 21 is a cross-sectional view of a paper surface when unevenness in time difference occurs

FIG. 22 is an explanatory diagram of a time difference unevenness factor

FIG. 23 is an explanatory diagram of dot misalignment

FIG. 24 is a diagram showing correspondence of dot deviation according to the third embodiment of the present invention.

FIG. 25 is a diagram illustrating a dot shift correspondence diagram of the second embodiment of the present invention.

FIG. 26 is a diagram illustrating a conventional dot shift correspondence diagram.

FIG. 27 is a dot array diagram of the first embodiment.

FIG. 28 is a reference diagram of the dot array of the third embodiment.

FIG. 29 is a reference diagram of the dot array of the third embodiment.

FIG. 30: Correspondence between the first embodiment of the present invention and the dither method

FIG. 31 is a comparison diagram of two dither methods.

[Explanation of symbols]

 701 4-color head 702 Multi-head 703 Paper feed roller 704 Auxiliary roller 705 Paper feed roller 706 Carriage 707 Printing paper

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification code Internal reference number FI Technical display location B41M 5/00 A 8808-2H B41J 3/10 101 E

Claims (4)

[Claims] 1. A print image is printed on a paper surface by performing a reciprocal recording scan a plurality of times for the same image area using a multi-head in which a plurality of ink ejection ports are arranged and relatively sequentially feeding the paper. In the ink jet printing apparatus for completing the above, the divided thinning pixel array in each printing scan is in a complementary relationship, and the divided thinned pixel array is a rectangular basic of m <n consisting of vertical m pixels and horizontal n pixels. An ink jet recording apparatus comprising a divided print control unit in which pixel groups are arranged according to a predetermined arrangement rule. 2. m <n comprising the vertical m pixels and the horizontal n pixels
Of the rectangular basic pixel group, the number of pixels p continuously arranged in the vertical direction is at least m <p <n.
An array rule satisfying the above condition.
The inkjet recording device described. 3. A print image is printed on a paper surface by using a multi-head having a plurality of ink ejection openings arranged, performing reciprocal recording scans for the same image area a plurality of times and relatively sequentially feeding the paper. In the ink jet printing method for completing the above, the divided thinning pixel array in each printing scan has a complementary relationship, and the divided thinning pixel array is a rectangular basic of m <n consisting of vertical m pixels and horizontal n pixels. An inkjet recording method characterized in that pixel groups are arranged according to a predetermined arrangement rule. 4. m <n consisting of the vertical m pixels and horizontal n pixels
Of the rectangular basic pixel group, the number of pixels p continuously arranged in the vertical direction is at least m <p <n.
4. An arrangement rule that satisfies
Inkjet recording method described.