JP2004175006A - Recording device, recording system, recording method - Google Patents

Abstract

[PROBLEMS] Calibration of line irregularities is performed by a printer main body. The processing can be performed by hardware as much as possible, and high-speed printing in the one-pass mode is realized while performing the streaking calibration.
A process is performed by scanning quantized image data for each pixel. If the pixel to be processed has print data, a nozzle correction value corresponding to the nozzle for printing the data is obtained, and an error from a predetermined surrounding pixel is added thereto. If the value is smaller than a predetermined threshold value, the data is changed so that the pixel is formed with a relatively low density. Thereafter, a determination error is calculated and assigned to a predetermined pixel.
[Selection diagram] Fig. 1

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a recording apparatus and a recording method for performing a correction process on image data. According to the present invention, various types of recording heads having a plurality of recording elements can be used as a recording head used for recording an image. In particular, an inkjet recording head in which a plurality of ink ejection units are arranged Alternatively, a thermal transfer recording head in which a plurality of thermal elements are arranged can be suitably used.
[0002]
[Prior art]
At present, recording methods include, for example, a thermal transfer method in which ink on an ink ribbon is transferred to a recording medium such as paper by thermal energy, and ink jet recording in which flying droplets are attached to a recording medium such as paper to perform recording. Methods are known.
[0003]
Among these, the ink jet recording method is widely used in printers and copiers because of low noise, low running cost, downsizing of the apparatus and easy realization of color. In general, a recording apparatus using such an ink jet recording system uses a recording head in which a plurality of recording elements are integrated and arranged in order to improve a recording speed. The recording element includes, for example, a nozzle for discharging ink and an ink discharge port.
[0004]
In such an ink jet recording apparatus, in the case of a serial scan system in which the recording head scans in the main scanning direction, one of the factors of image quality deterioration is recording unevenness (hereinafter referred to as "streak") which appears in a streak along the main scanning direction. Unevenness). The stripes often appear periodically, in which case they are very noticeable. For example, in a recording head of a so-called multi-nozzle type in which ink ejection ports are used, in order to eject ink from each ejection port, a heating heater (an electrothermal converter) located in an ink flow path communicating with each ejection port. In the case of using the heat generated by the heat generation, the following causes of unevenness can be cited. That is, variations in the ejection amount and ejection direction of the ink due to variations in the size of the heating heater and the ejection port in the nozzle unit, and the transport amount (paper feed amount) of the recording medium and the recording in the case of the serial scan method. The deviation from the width, the difference in the change in the density of the ink caused by the deviation in the recording time, the movement of the ink on the recording medium, and the like cause the occurrence of stripe unevenness.
[0005]
Conventionally, various methods have been proposed for achieving high image quality by eliminating such stripes.
[0006]
For example, in Japanese Patent Publication No. 59-31949, when a recording head repeatedly scans in the main scanning direction to record an image for each line in a serial scan system, a joint of recording areas for each line is provided. A method for preventing streaks from being generated in a portion is described. That is, the lowermost end of the recording area for the previous one line and the uppermost end of the recording area for the next one line are overlapped with each other. To complete the image.
[0007]
Also, as a conventionally known method for improving image quality, there is a division printing method (multi-pass printing method) in which printing in one printing area on a printing medium is completed by scanning the printing head a plurality of times. . Such a division recording method is effective in eliminating the occurrence of line unevenness. However, in order to sufficiently enhance the effect, the number of scans of the print head for one print area, that is, the number of divisions must be increased, and the throughput may be reduced accordingly.
[0008]
In any of such conventional methods, the print area completed each time the print head is scanned one time becomes small, and the throughput is reduced.
[0009]
As another method for suppressing the occurrence of line unevenness without using the divided recording method, for example, there is a head shading method as described in JP-A-5-69545. In this method, first, a test pattern for determining a correction value set in advance is recorded on a recording medium using a recording head, and the recording density of the recorded test pattern is read by a scanner. After appropriately correcting the position of the read image, the density of the image is assigned to a raster corresponding to each nozzle of the recording head. The change in the recording density is caused by a shift in the ink discharge amount or the ink discharge direction for each nozzle, or bleeding of the ink on the recording medium. Next, a correction value of the recording density corresponding to each nozzle is determined from the density data assigned to each raster. Then, based on the correction value, the gamma table for each nozzle is changed or the drive table for each nozzle is changed to change the ink ejection amount and the like. By such a correction, a raster recorded densely in the state without correction is corrected so as to be thin, and a raster recorded thinly in the state without correction is corrected so as to be darkened. Thus, the unevenness of the recording density is reduced.
[0010]
[Problems to be solved by the invention]
However, performing the conventional head shading has the following disadvantages.
[0011]
FIG. 34 is a flowchart schematically showing a data flow of a general printing apparatus that outputs a high-definition image. The input image data is first subjected to color processing for reproducing a preferable color. After that, a quantization process is performed. This process is, for example, a process of converting to a binary, ternary, or data format suitable for the configuration of the printing unit. Then, printing, that is, image formation is performed using the data. This processing is also the first two, namely, color processing and quantization are called an image processing unit that converts image data, and the subsequent printing is performed by physically moving a recording head or the like to perform printing and controlling data for that. It will be referred to as a print unit in the above.
[0012]
As one configuration of a system for forming an image by connecting a recording device (printer) to a host such as a computer, as shown in FIG. 35, the image processing unit is performed by a host computer, and data processed up to quantization is processed. Is transferred to a printer, and the printer forms (prints) an image.
[0013]
In the recording system having such a configuration, when the print data before quantization such as γ processing is processed by the head shading described in the conventional example to correct the stripes, a host computer is used as shown in FIG. This process needs to be performed on the side.
[0014]
On the other hand, in recent years, printing apparatuses are required to be able to perform high-speed printing, and therefore, the manner of printing and scanning of a printing head is changed according to the arrangement of images. As a specific example, in a printing apparatus having a relatively large number of nozzles and a long black nozzle row and a relatively small number of nozzles and a short color nozzle row, only a black image portion of an image to be printed is provided. Prints with the full use of the relatively long black nozzle row, and then feeds the paper the length of that nozzle. If the image to be printed has a portion to be printed in color, printing is performed using a relatively short color nozzle array, and the upper feed amount at that time is the width of the relatively short color nozzle.
[0015]
Further, as the recording speed increases, the temperature of the recording head rises sharply, and it is detected or estimated that the recording head is at a high temperature to change the print scanning method of the recording head.
[0016]
The correspondence between the data and which nozzle of the print head prints the data changes depending on the control factors of the printing unit. In a recording system in which the above-described processing of the image processing unit is performed by the host computer, it is necessary to spool the image processing data to check data in the image and to receive operation information from the printing unit. The former is very disadvantageous in terms of data processing speed because it is performed by computer software processing. The latter requires a communication means for notifying the host program of the printer data. In this case, the processing of the image processing unit in the host computer and the processing of the recording unit in the printer must be performed in cooperation with each other, so that it is difficult for the recording system to realize a high speed.
[0017]
As another method, if the correspondence between the image data and the nozzles of the recording head is simply determined in the streaking correction, and the printing unit is configured to form an image in accordance with that, the operation and control of the printing unit will be The processing for high-speed printing described above, that is, processing such as changing the operation of the recording unit in accordance with the image data and the temperature of the recording head, cannot be performed because of being restricted by the line unevenness control. The degree of freedom decreases.
[0018]
As another configuration of the recording system, there is a configuration in which the processing of the recording unit is also performed on the printer side. In this configuration, the case of performing the head shading described in the conventional example will be described with reference to FIG. In this case, there is no need to send the information of the print unit to the host computer. However, this case also has the following problems.
[0019]
When the operation of the printing unit is stopped, the speed of the recording device is reduced. Therefore, the CPU (central processing circuit) of the printer is normally used by the printing unit with priority. The data processing of the printing unit is performed prior to the printing process when the printing unit is not using the CPU. However, if the image processing unit performs the streaking correction process before the quantization process to perform the head shading, it is necessary to pass information of the printing unit to the image processing unit, and the image processing unit and the printing unit cooperate. Operation, which is disadvantageous for increasing the speed of the recording apparatus.
[0020]
Further, in the method of performing the correction process by changing the driving condition of the recording head, such as the method of changing the drive table for each nozzle to change the ink ejection amount described in the conventional example, the range of the density that can be changed, There is a limit on the number of correction levels that can be secured as a correction amount. This is because there is a physical restriction due to the mechanism of the recording head.
[0021]
An object of the present invention is to solve such a problem, to enable the image processing unit and the printing unit to operate independently as much as possible, and to realize high-speed printing while correcting line unevenness.
[0022]
It is another object of the present invention to secure a dynamic range of the correction amount and the number of correction levels sufficient for the correction process, and to form a higher quality image by performing the stripe unevenness correction process.
[0023]
[Means for Solving the Problems]
The present invention achieves the object by the following configurations.
[0024]
If there is data to be printed for the pixel of the image data after quantization, for the pixel of the image data, a value indicating the image forming characteristic of the nozzle row portion or nozzle of the print head corresponding to the pixel and a predetermined surrounding value A determination unit that determines a determination value using the error value received from the pixel, compares the determination value with a predetermined threshold value, and determines whether to control the image formation density of the pixel, and an image after quantization. If there is data to be printed on the pixels of the data, and if there is data to be printed for the pixels of the quantized image data, if there is data processing means for changing the processing method of the pixel data based on the result of the determination, An error value is calculated using a predetermined value according to a predetermined value from the determination value, and when there is no data to be printed for a pixel of the quantized image data, a predetermined surrounding value is determined. An error value determining means and said error values it the error value received from the element providing an error distribution means for distributing the predetermined pixel.
[0025]
With this configuration, the processing of the image data used for the nozzles in which the stripes occur is performed. That is, a value (hereinafter referred to as a nozzle correction value) indicating a portion of a nozzle row or an image forming characteristic of a nozzle is referred to as a nozzle correction value. The image data is changed or generated by determining whether to form an image at a relatively low density or to form an image at a relatively high density on the basis of an error.
[0026]
By this processing, it is possible to perform the stripe unevenness correction processing at the timing immediately before the image formation (printing), so that the system configuration is simplified, and it is not necessary to feed back the information of the printing unit to the image processing unit. Each process can be performed using time effectively, and a smoothing process can be performed at high speed.
[0027]
BEST MODE FOR CARRYING OUT THE INVENTION
(System configuration)
In order to perform the correction of the stripe unevenness, it is necessary to have a function of roughly acquiring the correction information of the stripe unevenness and a function of performing the processing of the stripe unevenness correction using the information, but the present invention proposes The latter. As the means for acquiring the former line unevenness information, data acquisition by a scanner as in a conventional head shading method (for example, JP-A-5-69545) may be performed. Further, density unevenness data may be acquired using a density sensor proposed in reference number 4033056. In still another example, at the time of manufacture of the printhead, the characteristics of the printhead may be measured and stored in a printing means provided on the printhead, and this may be used.
[0028]
In the present invention, as shown in FIG. 1, after the color processing and the quantization processing, the unevenness correction processing is performed by processing the quantized image data using the information of the printing mechanism. As a result, it is possible to perform the correction processing of the unevenness immediately before printing. Examples of the information of the printing mechanism include information on changing the printing method corresponding to the arrangement of the images described above and changing the printing method and conditions depending on the temperature of the recording head.
[0029]
As a result, it is possible to perform the printing and the unevenness correction processing of FIG. 1 without synchronizing with the image processing unit having a heavy calculation load, such as color processing and quantization processing. Of course, since the image-processed data is printed, the processing of the image processing section must be performed first.
[0030]
Originally, the streaking phenomenon corrects the characteristics of the recording apparatus. Therefore, the streaking correction processing is performed before printing, and a system configuration to be included in the printing unit is used. Thus, there is no need to provide feedback, and a simple configuration can be achieved. As described above, this has a great advantage in increasing the speed of the recording apparatus, and at the same time, more characteristics of the printing mechanism can be fed back to the stripe unevenness correction processing immediately before printing, so that the correction processing can be performed. The degree of freedom in design increases.
[0031]
Furthermore, the configuration of the present invention has another merit when performing a printer control method that changes the printing method in accordance with the arrangement of images while performing line unevenness correction. In order to determine how the image portion to be maintained is arranged in the input image data, it is necessary to store the image data in some form and check the image data. In that case, the data size of the quantized image data is generally smaller. Therefore, the memory for temporarily storing the image and the like can be reduced, and the time required for the process of checking the arrangement of the image can be shortened.
[0032]
The configuration of the present invention has a particularly great advantage in a recording system in which the image processing unit is performed by the host computer, and the processing of the printing unit and the image formation are performed by the printer. This is because sending the information of the printing unit to the host computer makes it difficult to speed up the recording system and also greatly restricts the design of the stripe unevenness correction processing.
[0033]
Even in a recording system in which both the processing of the image processing unit and the processing of the stripe unevenness control unit are performed by the print unit, there are significant advantages as described above in increasing the speed and increasing the degree of freedom in the design of the stripe unevenness correction. is there. For example, the processing of the image processing unit can proceed ahead of the idle time of the processing of the printing unit, which relatively preferentially uses the central processing circuit (CPU) of the recording apparatus. Can be used efficiently and the speed can be increased.
[0034]
(Dynamic range of line unevenness correction processing)
In the present invention, as described above, after the quantization processing, the unevenness correction processing is performed by processing the quantized image data using the information of the printing mechanism. More specifically, with respect to image data to be formed thinner, image data selected by a predetermined method (described later) is changed so that the density of image formation is lower than the original. Alternatively, the data level of the pixel is lowered. For example, the level is lowered from level 1 to level 0 or from level 2 to level 1.
[0035]
This method has the merit of handling the quantized data, and has the merit described above regarding the system configuration, and also has the merit that the dynamic range of the density that can be changed by the streaking process is large. Since the image data is changed by selecting a dot, it is also possible to omit all the image data for the thinner one.
[0036]
(Technical issues when processing quantized data to perform streaking processing)
The technical issues when processing the quantized data to perform streaks are: (1) to prevent unnatural dots or spaces from being clustered and to prevent textures from being formed; (2) To speed up the stripe unevenness correction processing itself.
[0037]
(Streak unevenness correction processing)
The stripe unevenness correction processing of the present invention includes, at least for each of the nozzles in the nozzle row of the print head, or for a plurality of nozzles, whether or not to make the image formation with the nozzles thinner, or how thinly it is, to the stripe unevenness correction data The processing is performed based on the selected pixel, and a pixel for lowering the density of image formation is selected, and the image data of the pixel is changed.
[0038]
Further, the following operation is performed in order to prevent the above-mentioned unnatural dot or space cluster (lump) or texture of (1).
[0039]
If there is data to be printed in a pixel for processing the quantized image data, a multivalued value is assigned based on a nozzle correction value obtained in advance.
[0040]
・ Add errors from surrounding pixels.
[0041]
Determine whether to lower the pixel density by comparing with a predetermined threshold value.
[0042]
・ Calculate the error generated by the determination.
[0043]
-Distribute the error to predetermined surrounding pixels.
[0044]
If there is no data to be printed at the pixel on which the quantized image data is to be processed, an error from the surrounding pixels is obtained and distributed to predetermined surrounding pixels.
[0045]
As for the pixels to which the error is distributed, the error is distributed to at least one pixel of the pixel of the data line to be processed next, as seen from the data line of the pixel in the scanning direction of the line unevenness processing and the other pixels being processed.
[0046]
Hereinafter, a more specific description will be given while showing an example with reference to the drawings. FIG. 2 is a schematic diagram for explaining the principle of the line unevenness processing of the present embodiment. A means for acquiring the correction information of the unevenness will be described in a later embodiment. A schematic diagram of the recording head is on the left side of FIG. To the right is a nozzle correction value corresponding to the nozzle. This is an integer value of a minimum of 0 and a maximum of 31. If it is desired to make the image formed by the nozzle of the corresponding print head thinner, a smaller value is assigned. On the right side of FIG. 2, there is a schematic diagram of the image data. Pixels in which a gray circle is drawn are pixels having data to be printed. This image data is binary data.
[0047]
Further, the printer of this example is capable of forming large dots and small dots on a recording medium, and forms an image with all large dots unless the stripe unevenness correction processing is performed. By replacing the large dots with small dots in accordance with the nozzle correction value by the stripe unevenness correction processing, the image data is changed so that the stripe unevenness becomes invisible.
[0048]
The determination as to whether to replace a large dot with a small dot is made by replacing the large dot data with the small dot data if the value obtained by adding the error from the surrounding pixels to the nozzle correction value corresponding to the nozzle is smaller than a predetermined threshold value. Change to If not, the large dot data is left as it is. In this example, the threshold for this determination is 21.
[0049]
If the error did not reduce the image formation density of a certain pixel of the print data, that is, if the data for printing a large dot was not changed, the corresponding nozzle correction value and the error from the surrounding pixels were added. Calculate by subtracting 31 from the value. When the pixel density is reduced, that is, when data for printing large dots is changed to data for printing small dots, 0 is subtracted from the value obtained by adding the error from the corresponding nozzle correction value and the surrounding pixels. Calculate.
[0050]
Then, the generated error is distributed to a pixel adjacent to the processing progress direction and a pixel in a direction perpendicular to that direction and next to a column on which the line unevenness processing is to be performed next by １ ／. The value after the decimal point is rounded down to calculate by bit shift.
[0051]
FIG. 3 shows a state in which the stripe unevenness correction processing has begun. The processing is performed downward from the upper left pixel in the image data in the figure. In FIG. 3, the pixels on which the stripe unevenness processing is performed are displayed in gray. There is no data to be printed in this pixel (the data to be printed is depicted by a gray circle in the figure). Also, there are no errors assigned from other pixels. The remaining error is zero. Next, the error is distributed to the pixel in the column in which the error is to be processed and to the column on which the stripe unevenness process is to be performed next, and the error is divided by １ ／.
[0052]
In the next step (FIG. 4), one pixel in the figure is processed. Since this pixel has data to be printed (shown by a gray circle in the image data image of FIG. 2), a value of 25 is obtained by referring to the nozzle correction value of the nozzle that prints this data. An error from the previous pixel is added to this, and the value of the error is 0, so the added value is also 25. This is compared with the change of the data of the large dot to that of the small dot or with the synchronization threshold, that is, 21 in this example. Since it is larger than the threshold, this pixel is left as a large dot and the data is not changed. When a new error generated by this determination is calculated, 31 is subtracted from 25 to be −6. This error is assigned to a pixel adjacent in the processing direction and a pixel in a direction perpendicular to the pixel in a half direction. In this case, it is -3. (See Fig. 4)
In the next step (FIG. 5), the nozzle correction value is 18, and the error from surrounding pixels is -3. Since the added value 15 is smaller than the judgment threshold value 21, the data of this pixel is changed to small dot data. The error is +15 by subtracting 0 from 15. An error of +7 is distributed to a pixel which has not been processed in the direction in which the processing proceeds and in a direction perpendicular to the direction in which the processing proceeds, by halving by bit shift.
[0053]
In the next step (FIG. 6), the nozzle correction value is 18, and the error from surrounding pixels is +7. Since the added value 25 is equal to or larger than the threshold value 21 for determination, this pixel remains a large dot and the data is not changed. The error is -6 minus 31 minus -6. The error -3 is assigned to the pixel which has not been processed in the direction in which the processing proceeds and in the direction perpendicular to the processing progress direction by halving by bit shift.
[0054]
In the next step (FIG. 7), since there is no print data, only errors allocated from surrounding pixels are handled. The error is -3. It is halved by a bit shift, and is distributed to pixels in which processing has not been performed in the direction in which the processing proceeds and in the direction perpendicular thereto. The sorting error is -1.
[0055]
Further, there is no stamp data in the next step (FIG. 8), and errors from surrounding pixels are sorted. The sorting error is zero.
[0056]
FIG. 9 shows the state five steps ahead, and FIG. 10 shows the state of the next step further. Pixels displayed in gray are pixels being processed. When the processing of one column of pixels is completed, the processing of the next column is performed in the opposite direction. This is to prevent errors from accumulating in either direction. In the step of FIG. 10, there is no print data, and the error from the surrounding pixels is -1. Therefore, the error of 0 is distributed to the direction in which the processing proceeds and to the pixels in the direction perpendicular to it and not yet processed. Since an error of -3 has already been distributed to the pixels in the traveling direction, the error 0 generated in this step is added, and the error of -3 remains.
[0057]
Proceed to the next step (FIG. 11). Since this pixel has print data, an error from surrounding pixels is added to the nozzle correction value. It is 15 by adding 18 and -3. Since this value is smaller than the judgment threshold value 21, the data of this pixel is changed to data of a small dot. The generated error is +15 by subtracting 0 from 15. It is halved by a bit shift, and is distributed to pixels in which processing has not been performed in the processing progress direction and in a direction perpendicular thereto. The sorting error is +7. Since there is already an error of +7 in the pixels adjacent in the processing direction, the addition results in an error of +14.
[0058]
In the next step (FIG. 12), the nozzle correction value is 18, and the error from surrounding pixels is +14. Since the added value 32 is equal to or larger than the judgment threshold value 21, the pixel remains a large dot and the data is not changed. The error is 1 minus 32 minus 31. An error of 0 is allocated to the pixel which has not been processed in the direction in which the processing proceeds and in the direction perpendicular to the direction in which the processing proceeds, by ビ ッ ト by bit shift. Since there is already an error of -3 in the pixels adjacent in the processing direction, the addition results in an error of -3.
[0059]
In the next step (FIG. 13), the nozzle correction value is 25, and the error from surrounding pixels is -3. Since the added value 22 is equal to or larger than the determination threshold value 21, this pixel remains a large dot and the data is not changed. The error is -9 minus 22 minus 31. An error -4 is assigned to a pixel that has not been processed in the direction in which the processing proceeds and in a direction perpendicular to the processing in the direction halved by bit shift.
[0060]
In this way, it is possible to change the large dot data into small dots by reflecting the nozzle correction value, and control the density of image formation.
[0061]
One of the important points of this method proposed by the present invention is that the error is distributed to pixels in the processing direction and pixels in the data string to be processed next in other directions. This makes it possible to arrange pixels or dots for lowering the density of image formation as uniformly as possible while controlling the number of such pixels or dots, while reflecting the arrangement of data to be originally printed. This prevents pixels with reduced density from appearing as thin clusters (clusters) and preventing lowering of printed products. Also, the texture is less likely to appear.
[0062]
As a method of selecting a pixel for lowering the density of image formation, a method of performing an AND operation of a fixed pattern and image data according to a nozzle correction value was also tried. However, the number of dots to be selected varies depending on the arrangement of the image data, and the processing result is obtained. The image density could not be controlled sufficiently.
[0063]
Further, the stripe unevenness processing is performed in a direction perpendicular to the arrangement of the nozzles of the recording head, that is, in a direction perpendicular to the example shown above, and the print data is counted, and a predetermined ratio according to the nozzle correction value is calculated. I also tried to select a pixel to lower the density of image formation with. That is, if the stripe unevenness processing is performed in a direction perpendicular to the arrangement of the nozzles of the recording head, the stripe unevenness processing for that row may be performed by reflecting only one nozzle correction value. Therefore, a process of counting data to be printed in the image data of the row and selecting a predetermined ratio, for example, one out of four data to be printed, and changing the data so that the density of image formation of the pixel becomes low. It can be performed.
[0064]
However, this method can control the number of pixels whose data is changed, but cannot control the arrangement of the pixels. When the stripe unevenness processing is performed, it is not possible to reflect the arrangement of the pixels whose data has been changed in other image data columns. As a result, in the direction perpendicular to the direction in which streaking processing is performed, pixels whose data has been changed are spatially solidified, and pixels whose data is not changed are spatially solidified, so that clusters and textures can be seen and print quality can be reduced. It got worse.
[0065]
Based on these experimental results, in order to perform streaking processing while minimizing clusters and textures, it is important not only to determine whether the data has been changed, but also to determine whether the data has been processed. We arrived at the conclusion that it was essential to be reflected in the processing, and devised the processing of the present invention.
[0066]
In the above example, the error is calculated from the pixel being processed, the pixel adjacent to the processing direction, and the pixel adjacent to the direction perpendicular to that direction and in the direction of the next data to be processed. The error was divided by 1/2. FIG. 14A shows the method of distribution. The present invention is not limited to this configuration. For example, the pixels may be divided into three pixels as shown in FIG. In some cases, it is more advantageous to allocate more pixels to prevent texture. However, in terms of speeding up the stripe unevenness correction processing described later, it is better that the number of processing steps is small, and in that regard, it is more advantageous to allocate the pixels to two pixels.
[0067]
Another point of the line unevenness processing proposed by the present invention is to convert image data once quantized into multi-valued data having more levels. Thus, the nozzle correction value can be specified in the step of the multi-value level after the conversion. This secures the gradation of the stripe unevenness correction processing, and makes it possible to correct subtle streaks and unevenness.
[0068]
Some recording systems perform quantization processing at 16 levels or the like, develop the data into binary data in a printing unit, and perform printing. Also in the recording system having such a configuration, the same effect can be obtained by performing the processing described above after developing the binary data in the printing unit.
[0069]
Still another point of the stripe unevenness processing proposed by the present invention is how to lower the density of pixels of image data. In order to reduce the density of a certain portion of the image for the processing of the stripe unevenness correction processing, it is simplest to thin out image data. If the dots formed by the device that listens on the recording medium are sufficiently small or thin and are not conspicuous, the print quality does not drop much even if thinned out. However, when the dots are relatively large, if the dots are thinned out, a space is created in the dots, which may be noticeable and degrade the image quality. The one where the dots stand out is called “dot power”. Then, it can be said that when the power of the dot is large, if the dots are thinned out in the stripe unevenness correction processing, the print quality may be degraded.
[0070]
Therefore, in the present invention, when performing the stripe unevenness processing, if the printing apparatus of the printing system can form dots of a plurality of powers, in order to obtain the highest possible image quality while correcting the stripe unevenness, the processing target The image data is changed so that the density of the image is reduced without changing the power of the pixel as much as possible. In the previous example, large dots were changed to small dots. As a result, small dots are formed at positions other than the large dots by the stripe unevenness correction process, and the small dots are less noticeable. Therefore, the density of the formed image can be controlled without lowering the print quality as much as possible.
[0071]
The processing of the present invention is not limited to changing the above-described large dot data to small dot data. It is only necessary that the density of the pixel can be reduced without changing the power as much as possible. For example, in a printing apparatus in which one pixel is formed by a plurality of dots, for a pixel to be thinned by the streaking process, the dot forming the pixel is used. May be reduced. As described above, the image processing unit may quantize the data, and the data sent to the printing unit may be developed into binary data by the printing unit, and then the line unevenness processing may be performed.
[0072]
Of course, in a recording system capable of forming dots that are sufficiently small, even if the dots are thinned out, it is not noticeable, so the dots may be simply thinned out.
[0073]
Further, in the present invention, in the streaking correction processing, a threshold value for determining whether or not to change the image data so as to form a relatively thin image is devised. In the above example, the nozzle correction value can take 32 levels from 0 to 31. However, the threshold value for the above determination is set to a higher level of 21 instead of 15 near the middle of the possible range. did. When the threshold value for determination is high, a determination to change data so as to lower the density of image formation occurs earlier. If the threshold value for the determination is set to a low value, the determination of the data conversion is unlikely to occur at first. Since a negative error gradually accumulates and a determination of data change is made, a portion that is not changed is formed at the beginning, which may cause a false contour of an image and deteriorate image quality.
[0074]
In fact, if the nozzle correction value is set at about the middle of a possible range such as 15, for example, in the example described above, the change of data from a large dot to a small dot occurs with a probability of about 50%. If it is not necessary to reduce the density to that extent in order to correct streaking unevenness, the nozzle correction value will be distributed at a higher place than the middle value of 15. In this case, if the threshold value for the determination is much smaller than those nozzle correction values, the determination of the conversion of the data change is unlikely to occur at the beginning of the line unevenness correction, and the possibility of occurrence of a false contour increases. . In the above-described example, in consideration of this, the threshold value is set to 21 which is higher than the intermediate value of 15.
[0075]
Further, the threshold of the determination may be set by checking the distribution of the nozzle correction values. For example, when the distribution of the nozzle correction values is distributed at a position higher than the center of the levels that the correction values can take, the nozzle correction value may be set to a level slightly lower than the lowest level, for example, two levels lower. . That is, in this case, the ratio is set slightly lower in consideration of the fact that the ratio of the pixels for which the data change is to be performed will be smaller than the ratio of the pixels for which the data change should not be performed. When the minimum level of the nozzle correction value is lower than the intermediate level, the threshold value for determination may be set to the intermediate level 15. This is to prevent the threshold value for determination from becoming too low, and conversely, causing the data change to occur too much at first to cause adverse effects on the image.
[0076]
The present invention is not limited to a configuration in which data is not changed when it is determined that the density of a pixel is not reduced. As a result of the determination, it is only necessary that data for forming a relatively dark image and data for forming a relatively light image of a pixel undergoing the stripe unevenness correction process can be obtained as a result of the stripe unevenness processing. Therefore, for example, the print data may not be changed when it is determined to be lower than the threshold, but may be changed so that relatively dark pixels can be formed when it is determined to be higher than the threshold. In both cases, new data may be generated.
[0077]
FIG. 15 shows a flowchart of the stripe unevenness correction processing of the present invention. In this example, a configuration has been described in which new data is created regardless of whether the density of a pixel to be processed is reduced or not, and where there is no data to be printed.
[0078]
(High-speed streak correction processing)
Although it has been described that the system configuration of the present invention is advantageous for speeding up, if the processing speed of the actual stripe unevenness correction processing is slow, the advantage is lost. Therefore, in the present invention, some measures are taken to increase the speed.
[0079]
In order to realize such a function in a recording system, it is possible to perform processing that requires a long time by using hardware rather than performing all processing using software, so that processing can be performed at high speed. Of course, if time allows, it may be constituted only by software. For this purpose, the present invention has the following configuration.
[0080]
First, the number of levels (the number of gradations) of the nozzle correction value is set to be greater than the number of quantized levels and smaller than that of the original image. The number of levels of the nozzle correction value greatly affects the circuit scale and the number of calculations when this processing is implemented by hardware. This is because, as described in the description using the previous example, in the present invention, a large number of calculations are performed using the number of levels of the nozzle correction value. As the circuit size increases, the cost of the recording system increases, and as the number of calculation circuits increases, processing time increases. Therefore, the number of levels is set to be as small as possible and necessary to realize the gradation required for the stripe unevenness processing.
[0081]
In the configuration of the recording system, the gradation of the image data is lower than that before quantization by performing quantization. However, if processing is performed at the gradation level before quantization in the streaking process, the gradation It is guaranteed that the gradation of the image data being processed is not further reduced due to the number of levels.
[0082]
However, when hardware is used for the above-mentioned reason, the number of levels of the nozzle correction value that is frequently used in the calculation of the process is directly related to the cost and the processing time of the hardware, and therefore, depends on the image forming capability of the recording apparatus. The number of levels of the nozzle correction value is reduced to an acceptable level. As a result, the circuit scale of the hardware and the number of calculations can be reduced.
[0083]
Of course, if the circuit scale of the hardware or the processing speed is sufficiently fast, the number of levels (the number of gradations) may be the same as the number of levels before quantization.
[0084]
Next, the number of pixels on which errors are scattered was minimized. As described above, the case where the number of pixels for which an error is scattered is large may be advantageous in preventing texture and cluster. However, the number of destinations of the error affects the number of calculations, which affects the processing time. In this regard, as in the example shown above, the error is allocated to two pixels: a pixel adjacent to the processing direction and a pixel perpendicular to that direction and adjacent to the processing direction. Necessary minimum and advantageous. Texture The image quality can be sufficiently covered by the other measures described above.
[0085]
However, if there is room in the processing time or the hardware circuit scale, the error may be allocated to more than two destinations.
[0086]
Furthermore, in order to reduce the scale of the hardware for the stripe unevenness correction processing, the width of the band for which the stripe unevenness correction processing is performed at one time is shorter than the width of the image to be printed, and the band is divided into a plurality of times. This will be described with reference to FIG. In this method, a print scan is performed by the print head to perform printing (image formation) for one band, and thereafter, the print medium is moved by the width of the band. This is an example of a serial scan recording system that forms an image by repeating this. In the image data shown in the figure, a banding correction process is performed in band units having a width corresponding to the head length.
[0087]
According to this configuration, the hardware requires only a register for storing a number of nozzle correction values corresponding to the number of nozzles of the print head and a register for storing an error value. Compared to the case where the correction of the stripe unevenness for one page is performed at once, the scale of the hardware can be small, and the cost is also advantageous. Also, it is only necessary that the banding correction processing for the band of the image data has been completed before the printing scan of the data of the band of the image data is performed, and that the CPU time becomes idle until the printing scan of the data of the next band is performed. It is only necessary that the processing for the band to be used again has been completed. This method also has an advantage in that time is efficiently used. If the CPU has a free time, the banding correction processing may perform the processing of the data of the band further ahead.
[0088]
However, in this configuration, there is a possibility that a pseudo contour may occur between the bands, and it is necessary to design each specification of the banding process so that the image quality does not deteriorate.
[0089]
An example of another configuration is shown with reference to FIG. 17. In this case, the width of the band of the image data to be processed at one time is half the width of the recording head. In this case, before performing the banding correction processing for one band, a nozzle correction value is set according to the correspondence between the image data and the nozzles of the recording head, and then the nozzle correction processing is performed. This configuration has an advantage that the scale of hardware can be further reduced as compared with the example on the left. The width of one band is not limited to half the length of the recording head. Further, by reducing the width of the band in which the stripe unevenness correction processing is performed at one time, the scale of hardware for performing one stripe unevenness correction processing can be reduced. Instead, the total time required for the correction processing of the data unevenness required for one print scan is somewhat increased by calling the function of the hardware a plurality of times.
[0090]
In addition, since data processing for one print scan is performed a plurality of times, if there is a band of image data that does not require line unevenness processing, the processing can be passed.
[0091]
These configurations have advantages such as the capacity of a memory for holding error data even when the function of correcting unevenness is realized by software.
[0092]
An example of still another configuration will be described. FIG. 18 is a diagram in the case where the stripe unevenness processing is performed in the scanning direction of the recording head. This method requires a means for holding the data of the discrimination error for one row in the horizontal direction. However, in this configuration, there is no boundary between bands, which occurs when the above two image data are processed by being divided into a plurality of bands. Therefore, the risk of generating a false contour can be avoided.
[0093]
In the description of these three methods, the recording apparatus of the serial scan system has been described, but the present invention is not limited to the configuration. A recording system that performs recording using a recording head having a vertical or horizontal length on a recording medium has the same advantage.
[0094]
Although a single color print has been described above, the present invention is not limited to this configuration. In a printing apparatus that performs printing of a plurality of colors, the same effect can be obtained by performing the stripe unevenness correction processing on the print heads of each color and each image data.
[0095]
(First embodiment)
This embodiment will be described using a case of a single color for easy description.
[0096]
(Configuration of recording device)
FIG. 19 is a perspective view showing a printer as an inkjet printing apparatus according to the embodiment of the present invention.
[0097]
In FIG. 19, reference numeral 101 denotes a printer, 102 denotes an operation panel unit provided at the front of the upper surface of the housing of the printer 101, 103 denotes a paper feed cassette mounted through an opening on the front surface of the housing, and 104 denotes a paper feed cassette. A paper (recording medium) supplied from the paper feed cassette 3 is a paper discharge tray 105 for holding paper discharged through a paper transport path in the printer 101. Reference numeral 106 denotes a main body cover having an L-shaped cross section. The body cover 106 covers an opening 107 formed in the front right portion of the housing, and is rotatably attached to an inner end of the opening 107 by a hinge 108. A carriage 110 supported by a guide or the like (not shown) is provided inside the housing. The carriage 110 is provided so as to be able to reciprocate along a width direction of the paper passing through the paper transport path (hereinafter, also referred to as a main scanning direction).
[0098]
The carriage 110 in the present embodiment has a stage 110a that is horizontally held by a guide or the like, an opening (not shown) for mounting an inkjet head behind the stage 110a, and the opening 110 A cartridge garage 110b for accommodating the detachably mounted inkjet heads 3Y, 3M, 3C and 3Bk, and a cartridge holder 110c that is opened and closed with respect to the garage 110b to prevent the cartridge accommodated in the garage 110b from being detached. It is roughly constituted from.
[0099]
The stage 110a is slidably supported by a guide at a rear end thereof, and a lower side of a front end thereof is slidably engaged with a guide plate (not shown). The guide plate may function as a paper holding member for preventing the paper conveyed on the above-described paper conveyance path from rising, and the stage may be cantilevered relative to the guide according to the thickness of the paper. It may have a function of lifting the object.
[0100]
An ink jet head (not shown) is attached to the opening of the stage 110a with its ink ejection port facing downward.
[0101]
In the cartridge garage 110b, through holes are formed in the front-rear direction for accommodating the four ink cartridges 3Y, 3M, 3C, 3Bk at the same time, and engaging claws of the cartridge holder 110c are engaged on both outer sides. A coupling recess is formed.
[0102]
The cartridge holder 110c is rotatably attached to the front end of the stage 110a by a hinge 116. The dimension from the front end of the garage 110b to the hinge 116 is determined in consideration of, for example, the size of the cartridge 3Y, 3M, 3C, 3Bk projecting from the front end of the garage 110b when housed in the garage 110b. The cartridge holder 110c has a substantially rectangular plate shape. In the cartridge holder 110c, the upper part of the upper part distant from the lower part fixed by the hinge 116 projects in a direction perpendicular to the plate surface, and when the holder 110c is closed, the engagement recess 110d of the garage 110b is formed. A pair of engaging claws 110e to be engaged are provided. In the holder 110c, a fitting hole 120 for fitting the handle of each of the cartridges 3Y, 3M, 3C, 3Bk is formed in its plate. These fitting holes 120 have a position, a shape, and a size corresponding to the handle.
[0103]
The carriage 110 is provided with a scanner sensor (not shown). The image of the printed pattern can be read by this sensor.
[0104]
As shown in FIG. 20, two heaters SH1 and SH2 are arranged in the longitudinal direction of each ink path 2A. These two heaters have different surface areas from each other. Each heater can be driven separately and independently, and electrode wiring and the like (not shown) are provided so that the two heaters can be driven simultaneously. Is provided. Note that the heater SH1 and the heater SH2 have the same length in the longitudinal direction of the ink path 2A, and have different surface areas by changing their widths. An ejection port 2N is open at the tip of the ink path 2A.
[0105]
By driving the heater SH1 alone, ink droplets of a relatively small discharge amount are discharged. Further, by driving the heaters SH1 and SH2 simultaneously or at a predetermined timing, it is possible to discharge a relatively large amount of ink droplets. Details of the drive timing and the like are disclosed in JP-A-08-183179.
[0106]
(Acquisition of nozzle correction value)
In this embodiment, a nozzle characteristic measurement pattern having a duty of 50%, more specifically, a staggered pattern is formed on a recording medium by forward scanning and backward scanning, and is formed by a scanner sensor mounted on a carriage of the recording apparatus. It measures optically and calculates a nozzle correction value based on it.
[0107]
Also, for ease of explanation, a case will be described in which the recording head has eight nozzles. Of course, the invention is not limited by the number of such nozzles.
[0108]
First, the pattern for measuring the nozzle characteristics will be described with reference to FIG. First, as shown by (1) in the figure, a staggered pattern is formed on the recording medium by forward scanning. Thereafter, printing of a zigzag pattern is performed by forward scanning indicated by (2). Thereafter, the staggered pattern printing by the backward scanning indicated by (3) is performed, and finally the staggered pattern printing indicated by (4) is performed.
[0109]
Next, the nozzle characteristics are measured. In FIG. 22, a range measured by a scanner sensor is indicated by a dotted line on the nozzle characteristic measurement pattern formed on the recording medium. After measuring this range, averaging is performed in the horizontal direction to reduce noise. If the resolution of the scanner sensor and the resolution of the printhead are not the same, an interpolation process is performed to obtain a characteristic value for each nozzle of the printhead. If it is necessary to further reduce noise, moving average processing may be performed.
[0110]
FIG. 23 shows an example of measured values obtained for each of the forward scan and the backward scan. Hereinafter, this value is called a characteristic value. This figure shows that the larger the characteristic value, the higher the light intensity from the recording medium surface. That is, the higher the measured value, the lower the density of the portion corresponding to each nozzle of the formed pattern.
[0111]
The absolute value of the difference between the characteristic values corresponding to each nozzle is taken from the larger value of the characteristic values obtained for the nozzles. An example of the value is shown in FIG. A larger value (indicated by ΔI in the figure) indicates that the corresponding nozzle has a characteristic of forming a darker image.
[0112]
Data for converting this value (ΔI) and the nozzle correction value is prepared, and using the data, a nozzle correction value corresponding to each nozzle of the print head in the forward scan and the backward scan is obtained. The data for the data conversion physically means how much the nozzle correction value is required and how much the density can be corrected by the value of ΔI.
[0113]
(Streak unevenness correction processing)
In the first example, the line unevenness correction processing is performed in substantially the same manner as in the example described in the embodiment. That is, the following processing is performed.
[0114]
If there is data to be printed at a pixel for processing the quantized image data, a nozzle correction value corresponding to that pixel is obtained. Nozzle correction values range from 0 to 31.
[0115]
・ Add errors from surrounding pixels.
[0116]
Determine whether to lower the pixel density by comparing with a predetermined threshold value. The judgment threshold value is 21.
[0117]
・ Calculate the error generated by the determination. If the density of the pixel to be processed is reduced, 0 is subtracted, and if the density is not reduced, 31 is subtracted.
[0118]
The pixel for distributing the error to predetermined surrounding pixels is a pixel adjacent in the scanning direction of the stripe unevenness processing, and a pixel of a data string to be processed next in a direction perpendicular thereto.
[0119]
If there is no data to be printed at the pixel on which the processing of the quantized image data is to be performed, an error from the surrounding pixels is obtained, and a pixel adjacent to the scanning direction of the line unevenness processing and a direction perpendicular thereto, Further, it is a pixel of a data string to be processed next.
[0120]
This process is shown in the flowchart of FIG.
[0121]
In this embodiment, as shown in FIG. 16, the processing is performed in band units of the image data having the width of the recording head. Further, in this embodiment, an empty calculation is performed at the end of the band in order to prevent a pseudo contour from being generated between the bands.
[0122]
Hereinafter, the empty calculation will be described. FIG. 26 is a diagram for explaining the null calculation. The portion indicated by the solid line in the image data shown in the figure is the actual image data. Assuming that there is data one at a time above and below the width in the vertical direction corresponding to the nozzle width, the calculation of the stripe unevenness correction processing is performed. The presence / absence of image data and the nozzle correction values of the actual data inside are used.
[0123]
Hereinafter, the example of the stripe unevenness correction processing will be described in more detail. FIG. 27 is a diagram illustrating an example of a state in which the stripe unevenness processing of the one-band image data has been started. Assuming that image data for one pixel exists on both sides of the actual image data drawn by the solid line in the figure, the calculation of the line unevenness processing is performed. The processing is started from the upper left pixel of the image data in the figure and is displayed in gray. As described above, with respect to the pixel of the empty calculation, the nozzle correction value is used to determine whether there is actual data one inner side, that is, in this case, whether there is data to be printed for the lower pixel in the figure. That is, the calculation of this pixel is performed assuming that there is data to be printed and the nozzle correction value is 25. Since the error from the surrounding pixels has not been sorted out, 0 is added, and the value of this pixel remains at 25. This pixel is compared with the judgment threshold value 21 and is equal to or larger than the threshold value. However, since this pixel is a pixel assumed for sky calculation, the data of this pixel is not used for forming an image. An error generated by subtracting a predetermined value 31 from 25 is -6, which is 1/2 for a pixel adjacent to the processing progress direction and a pixel of a data string which is in a direction perpendicular thereto and which is to be processed next. An error of 2 (or a fraction of −3 is discarded in order to calculate by bit shift), that is, an error of −3 is assigned.
[0124]
Proceed to the next step. FIG. 28 is a diagram for explaining the stripe unevenness processing of the next pixel. The next pixel has data to be printed and the nozzle correction value is 25. The value obtained by adding the error of −3 allocated from the previous pixel is 22. This pixel is compared with the threshold value 22 for determination, and since the value is equal to or greater than the threshold value, this pixel also generates large dot print data. An error -9 is obtained by subtracting a predetermined value 31 when the dot is determined to be a large dot from the value 22. The error -4 whose value is shifted to ビ ッ ト is distributed to predetermined surrounding pixels.
[0125]
Proceed to the next step. This will be described with reference to FIG. This pixel has data to be printed and has a nozzle correction value of 25. A value of 21 is obtained by adding the error of −5 allocated from the previous pixel. This is compared with the threshold value 21 for determination, and since it is equal to or greater than the threshold value, this pixel also generates large dot print data. An error -10 is obtained by subtracting a predetermined value 31 determined to be a large dot from the value 21. An error -5 obtained by halving the value by a bit shift is distributed to predetermined surrounding pixels.
[0126]
Next, a state in which the processing is advanced by about six steps will be described with reference to FIG. Since there is no data to be printed for this pixel, the nozzle correction value is not used. An error of +9 assigned from the previous pixel is obtained, and an error +4 obtained by halving the value by bit shift is assigned to predetermined surrounding pixels.
[0127]
The next step will be described with reference to FIG. This pixel is a pixel assumed for the sky calculation, and refers to the nozzle correction value as to whether or not there is data to be printed in one pixel inside, that is, one pixel above in FIG. That is, the calculation is performed on the assumption that there is no data to be printed. An error +4 from the previously processed pixel is obtained, and the error +2 halved by bit shift is distributed to predetermined surrounding pixels. Since there is no pixel in the direction of processing, there is a direction perpendicular to that direction, and a +2 error is allocated to the pixel of the data string to be processed next, that is, the right pixel in the figure.
[0128]
From the next step, processing of a new data string is performed. As described in the above embodiment, the direction of the stripe unevenness processing is the opposite direction. This is to prevent errors from accumulating in one line. Since this pixel is also assumed for the sky calculation, there is data to be printed with reference to the upper pixel in FIG. The error +2 distributed from the surrounding pixels is added to the nozzle correction value, and the value becomes 27. By comparing with the threshold value of the determination, a determination of a large dot is obtained. However, since this pixel is assumed for the sky calculation, it is not used for forming an actual image. An error -4 is obtained by subtracting a predetermined value 31 in the case of determining a large dot from 27. This is assigned to a predetermined pixel by １ ／, that is, an error of −2. Since an error of +4 has already been assigned to pixels adjacent in the processing direction (upper pixel in the figure), if an error of -2 is further assigned, an error of +2 remains as a result of addition. .
[0129]
Further processing of the next pixel will be described with reference to FIG. In this process, there is data to be printed, and the correction value is 25. The value obtained by adding the error +2 distributed from the surrounding pixels to 25 is 27. This value is compared with the threshold value 21 for the determination. Since it is equal to or larger than the threshold value, large dot data is generated. An error is -4 by subtracting a predetermined value 31 in the case of determining a large dot from 27. This is assigned to a predetermined pixel by １ ／, that is, an error of −2. Since an error of +9 has already been assigned to pixels adjacent in the processing direction (upper pixel in the figure), if an error of -2 is further assigned, an error of +7 remains as a result of addition. . Hereinafter, the same processing is continued.
[0130]
Here, looking at FIG. 33, the fourth small dot appears including the sky calculation. The negative errors gradually accumulate after the start of the stripe unevenness correction processing and small dots appear, but large dots continue until then. Until such an error accumulates, the same determination continues at the band edge of the image data (upper and lower sides of the image data in the figure in the present embodiment), whereby a pseudo contour may be generated.
[0131]
However, in the image data used to form an image, data of a small dot appears third. That is, by performing the null calculation, the range where the same determination continues at the end is narrowed. This reduces false contours that occur.
[0132]
Also, in the end portion, the same determination or the determination in the same pattern tends to continue in the direction perpendicular to the direction in which the stripe unevenness correction process proceeds. The false contour due to this is also reduced by the sky calculation.
[0133]
Other effects of the configuration of this embodiment are not described here because they have already been described in the embodiment.
[0134]
(Second embodiment)
The second embodiment has almost the same configuration as that of the first embodiment. However, when an image of a pixel of image data is formed with a relatively high density, an image is formed with two dots per pixel. When pixels of image data are formed at a relatively low density, one dot is formed for each pixel.
[0135]
That is, in the flowchart of FIG. 15, when creating image data without lowering the density of the pixel, the image data is created so that two dots are formed in one pixel. When creating image data for lowering the density of a pixel, the image data is created so that one dot is formed for one pixel.
[0136]
The configuration of forming a dark or light image by controlling the number of dots forming one pixel into two pixels as in the present embodiment is different from the relatively large dot and the relatively small dot used in the first embodiment. The advantage over the method of forming pixels with dots is that the structure of the recording head and the control of the recording head are simplified. On the other hand, since it is necessary to drive the recording head that forms one pixel a plurality of times to form dots, it is disadvantageous with respect to the number of pixels that can be formed per unit time, that is, the speed of pixel image generation. is there.
[0137]
Although the above-described two image forming methods have been described in the first and second embodiments, the present invention is not limited to the configuration. It suffices if it is possible to control pixels that are formed darker and pixels that are formed lighter by changing or generating image data.
[0138]
(Third embodiment)
The third embodiment has almost the same configuration as the first embodiment. The difference from the first embodiment is that when an image of pixels of image data is formed with a relatively high density, the image is formed with one dot per pixel, and the pixels of the image data are formed with a relatively low density. If so, the pixel does not form a dot. That is, the dots are thinned out.
[0139]
In this configuration, the thinned dots are conspicuous as described in the embodiment, and the image quality may be lower than in the first and second embodiments. However, if the dots to be formed are small or thin and are difficult to see, even if the dots are thinned out, they become less noticeable and hardly cause an image product. An advantage of the method of the third embodiment is that control of the recording head and data can be performed simply.
[0140]
In this embodiment, as shown in FIG. 18, the unevenness correction control in the print scanning direction of the recording head is performed. The merit of this method has already been described in the embodiment, but there is no boundary between bands of image data as in the first embodiment. Thereby, the possibility that a false contour is generated there can be avoided.
[0141]
Further, in this embodiment, a nozzle correction value is assigned to every eight nozzles of the print head. This configuration is effective for correcting unevenness in which there is almost no line unevenness for each nozzle and the density of image formation gradually changes with respect to the nozzle row. This configuration has an advantage that the resolution for acquiring nozzle correction can be reduced. For example, the lower the resolution of the scanner head, the lower the cost of the scanner head.
[0142]
【The invention's effect】
As described above, according to the present invention, it is possible to perform the line unevenness correction processing at a timing immediately before forming (printing) an image, so that the system configuration is simplified and the information of the printing unit is stored in the image processing unit. Since there is no need to feed back to each other, each process can be performed using time effectively, and the line unevenness process can be performed at high speed.
[0143]
Also, in a recording system in which the image processing unit is processed by the host computer, the quantized image data is transferred to the printer, and the image is formed by the printer, the streaking process may be closed by the printer for processing. It becomes possible.
[0144]
Further, since the streaking correction processing is simplified and speeded up, processing can be performed at high speed using hardware having a relatively small circuit scale. This makes it possible to mount an inexpensive recording system and a recording apparatus to form an image at high speed while performing a stripe unevenness correction process.
[Brief description of the drawings]
FIG. 1 is a flowchart for explaining a processing flow of a recording system to which the present invention is applied.
FIG. 2 is a diagram for explaining the operation of a stripe unevenness correction process according to the present invention.
FIG. 3 is a diagram for explaining an operation of a stripe unevenness correction process according to the present invention.
FIG. 4 is a diagram for explaining the operation of the stripe unevenness correction processing of the present invention.
FIG. 5 is a diagram for explaining the operation of the stripe unevenness correction processing of the present invention.
FIG. 6 is a diagram for explaining the operation of the stripe unevenness correction processing of the present invention.
FIG. 7 is a diagram for explaining the operation of the stripe unevenness correction processing of the present invention.
FIG. 8 is a diagram for explaining the operation of the stripe unevenness correction processing of the present invention.
FIG. 9 is a diagram for explaining the operation of the stripe unevenness correction processing of the present invention.
FIG. 10 is a diagram for explaining the operation of the stripe unevenness correction processing of the present invention.
FIG. 11 is a diagram for explaining the operation of the stripe unevenness correction processing of the present invention.
FIG. 12 is a diagram for explaining the operation of the stripe unevenness correction processing of the present invention.
FIG. 13 is a diagram for explaining the operation of the stripe unevenness correction processing of the present invention.
FIG. 14 is a diagram for explaining a method of allocating an error according to the present invention.
FIG. 15 is a flowchart for explaining the stripe unevenness correction processing of the present invention.
FIG. 16 is a diagram for explaining how to perform scanning of processing performed on image data in the line unevenness correction processing of the present invention.
FIG. 17 is a view for explaining how to perform scanning of processing performed on image data in the line unevenness correction processing of the present invention.
FIG. 18 is a view for explaining how to perform scanning of processing performed on image data in the line unevenness correction processing of the present invention.
FIG. 19 is a perspective view illustrating an example of a mechanical configuration of the recording apparatus according to the first embodiment of the present invention.
FIG. 20 is a diagram illustrating an example of a configuration of a recording head according to the first embodiment of the present invention.
FIG. 21 is a diagram for explaining a nozzle characteristic measurement pattern according to the first embodiment of the present invention.
FIG. 22 is a diagram for explaining measurement of nozzle characteristics in the first embodiment of the present invention.
FIG. 23 is a diagram illustrating an example of a characteristic value according to the first embodiment of the present invention.
FIG. 24 is a diagram for explaining calculation of a nozzle correction value in the first embodiment of the present invention.
FIG. 25 is a diagram for explaining calculation of a nozzle correction value in the first embodiment of the present invention.
FIG. 26 is a diagram for explaining a stripe unevenness correction process according to the first embodiment of the present invention.
FIG. 27 is a diagram for explaining a stripe unevenness correction process in the first embodiment of the present invention.
FIG. 28 is a diagram for explaining a stripe unevenness correction process in the first embodiment of the present invention.
FIG. 29 is a diagram for explaining the stripe unevenness correction processing in the first embodiment of the present invention.
FIG. 30 is a diagram for describing a stripe unevenness correction process according to the first embodiment of the present invention.
FIG. 31 is a diagram for explaining a stripe unevenness correction process according to the first embodiment of the present invention.
FIG. 32 is a diagram for explaining a stripe unevenness correction process in the first embodiment of the present invention.
FIG. 33 is a diagram for describing a stripe unevenness correction process in the first embodiment of the present invention.
FIG. 34 is a flowchart for explaining a conventional example.
FIG. 35 is a flowchart for explaining a conventional example.
FIG. 36 is a flowchart for explaining a conventional example.
FIG. 37 is a flowchart for explaining a conventional example.

Claims (14)

Determining means for determining the control of image formation density of pixels with respect to quantized image data by using information indicating image forming characteristics for at least each part of the nozzle array of the recording head or for each nozzle; A data processing means for changing a data processing method. For a pixel of the quantized image data, image characteristic information acquiring means for acquiring a value indicating an image forming characteristic of a nozzle row portion or a nozzle of the recording head corresponding to the pixel, and information from predetermined surrounding pixels. Information obtaining means for obtaining, determining means for determining the control of the image formation density of the pixel using at least one of the information, and data processing means for changing the processing method of the pixel data based on the result. The recording device according to claim 1, further comprising: The information including the result of the determination for controlling the image formation density of the pixel is obtained by using the image of at least two pixels of the pixel in the processing direction of the data string being processed and the pixel of the data string to be processed next. 3. The recording apparatus according to claim 1, wherein the information is reflected in a determination for controlling the density of formation. Information including the result of the determination for controlling the image formation density of the pixel is stored in a pixel adjacent to the pixel being processed in the processing direction and in a data string to be processed next and perpendicular to the processing direction. 4. The recording apparatus according to claim 3, wherein the control unit reflects the image formation density of at least two pixels adjacent to each other in a predetermined direction. If there is data to be printed for the pixel of the image data after quantization, for the pixel of the image data, a value indicating the image forming characteristic of the nozzle row portion or nozzle of the print head corresponding to the pixel and a predetermined surrounding value Using the error value received from the pixel of
Determining means for comparing the determination value with a predetermined threshold value and determining to control the image formation density of the pixel;
Whether there is data to be printed on the pixels of the quantized image data or not, and data processing means for changing the processing method of the pixel data according to the result of the determination,
If there is data to be printed for the pixels of the quantized image data, calculate an error value using a predetermined value according to a predetermined value from the determination value,
When there is no data to be printed for the pixels of the quantized image data, an error value determination unit that receives from predetermined surrounding pixels and sets it as an error value,
5. The recording apparatus according to claim 2, further comprising: an error allocating unit configured to allocate the error value to a predetermined pixel. The data processing means, when there is no data to be printed on the pixel of the image data after quantization, so that no image is formed on the pixel,
If there is data to be printed in a pixel of the quantized image data, a relatively dark image is formed at that pixel according to the result of the determination means, or a relatively light image is formed in another determination. 6. The recording apparatus according to claim 1, wherein data is processed in such a manner as to be performed. The data processing means, when there is no data to be printed on the pixel of the image data after quantization, so that no image is formed on the pixel,
If there is data to be printed in a pixel of the quantized image data, a dot is formed at that pixel according to the result of the determination means, and in another determination, the data is processed so that the pixel is not formed. The recording apparatus according to claim 6, wherein: The data processing means, when there is no data to be printed on the pixel of the image data after quantization, so that no image is formed on the pixel,
If there is data to be printed in a pixel of the quantized image data, a relatively large dot is formed at that pixel according to the result of the determination means, and in another determination, the pixel is relatively large. The recording apparatus according to claim 6, wherein the data is processed so that small dots are formed. The data processing means, when there is no data to be printed on the pixel of the image data after quantization, so that no image is formed on the pixel,
If there is data to be printed in a pixel of the quantized image data, a relatively large number of dots are formed in that pixel based on the result of the determination means, and in another determination, the number of dots is relatively large in that pixel. 7. The recording apparatus according to claim 6, wherein the data is processed such that an extremely small number of dots are formed. The data processing means, when there is no data to be printed in the pixel of the image data after quantization, data in which no image is formed in the pixel,
If there is data to be printed at a pixel of the quantized image data, data of a relatively high level is given to the pixel according to the result of the determination means, and if another determination is made, data of a relatively high level is given to the pixel. 7. The recording device according to claim 6, wherein the recording device generates low-level data. 10. The print head according to claim 1, wherein a value indicating a nozzle row portion or a nozzle image forming characteristic has a greater number of tones than the number of tones of the quantized image data. The recording device according to the above. 11. The printing apparatus according to claim 10, wherein the value indicating the image forming characteristic of the nozzle row portion or the nozzle of the print head is smaller than the number of gradations of the image data before quantization. The direction in which the scanning of the image data is performed is a direction perpendicular to the scanning method of the recording head, and the width of the scanning in the processing is a width corresponding to the length of the nozzle row of the recording head. The recording apparatus according to claim 1, wherein the recording apparatus is characterized in that: The direction in which the scanning of the image data is performed is a direction perpendicular to the scanning direction of the recording head, and the width of the scanning in the processing is shorter than the width corresponding to the length of the nozzle row of the recording head. 14. The recording device according to claim 13, wherein the recording device is a number.

Images