JP4013855B2 - Image processing apparatus for converting image data in units of multiple pixels - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a technique for converting gradation image data into dot data expressed by the presence or absence of dot formation, and more specifically, image data is quickly converted by processing a predetermined number of adjacent pixels collectively. However, the present invention relates to a technique for avoiding deterioration in image quality.
[0002]
[Prior art]
An image display device that expresses an image by forming dots on a display medium such as a print medium or a liquid crystal screen is widely used as an output device of various image devices. Such an image display device can express only the state of whether or not to form dots locally, but by appropriately controlling the dot formation density according to the gradation value of the image, It is possible to express an image whose tone changes continuously.
[0003]
In these image display devices, a method called an error diffusion method is widely used as a method for determining the presence or absence of dot formation for each pixel so that dots are formed at an appropriate density according to the gradation value of the image. in use. In the error diffusion method, an error in gradation expression caused by the formation of a dot in a pixel of interest or the absence of a dot is diffused and stored in undetermined pixels around the pixel of interest. In determining the presence or absence of dot formation for a pixel, this is a method for determining the presence or absence of dot formation so as to eliminate errors diffused from surrounding pixels. By using the error diffusion method, it is possible to determine whether or not dots are formed so as to eliminate the error in gradation expression that has occurred in the peripheral pixels in this way, so that the density becomes appropriate according to the gradation value of the image. As described above, it is possible to determine the presence or absence of dot formation.
[0004]
By using the error diffusion method, high-quality images can be displayed by forming dots with an appropriate density according to the image. When the number of pixels constituting the image increases, it takes time to process and it is difficult to express the image quickly. In order to solve such a problem, the inventor of the present application has developed a technique for determining the presence or absence of dot formation while collecting a predetermined number of adjacent pixels in a unit and diffusing an error in units. (Patent Document 1).
[0005]
[Patent Document 1]
JP 2001-185789 A
[0006]
[Problems to be solved by the invention]
However, when determining whether or not dots are formed on a unit basis, the same processing is repeated for each unit. For this reason, a dot is formed in the pixel of the same position in a unit over a some unit, and this will be visually recognized as a periodic pattern, and there exists a possibility of degrading an image quality. Further, when the presence / absence of dot formation is determined on a unit basis, a problem that a periodic pattern may be generated may occur not only in the error diffusion method but also in other known methods.
[0007]
The present invention has been made to solve the above-described problems in the prior art, and it is possible to make a quick determination by determining the presence / absence of dot formation on a unit basis, while ensuring the generation of a periodic pattern. The object is to provide a technique capable of avoiding and displaying a high-quality image.
[0008]
[Means for solving the problems and their functions and effects]
In order to solve at least a part of the problems described above, the image processing apparatus of the present invention employs the following configuration. That is,
An image processing apparatus that converts image data represented by gradation values for each pixel constituting an image into dot data indicating the presence or absence of dot formation for each pixel,
Dot data conversion means for grouping a predetermined plurality of adjacent pixels as a unit, and converting the image data into the dot data in units.
Dot presence / absence storage means for storing the presence / absence of dot formation in a state where the pixel position in the unit can be identified for at least some of the pixels constituting the converted unit;
With
The dot data conversion means includes
A converted unit selection means for calculating a total sum of gradation values for each pixel constituting the unit, and selecting a predetermined number of units from the converted units according to a correspondence relationship with the total value;
A pixel-of-interest setting means for sequentially setting a pixel of interest for determining the presence or absence of dot formation in the unit to be converted into the dot data;
Dot formation presence / absence detection means for detecting the presence or absence of dot formation at a pixel at a predetermined position with respect to the pixel position of the pixel of interest for the selected predetermined number of units;
Dot formation determination means for determining the presence / absence of dot formation for the pixel of interest based on the gradation value of the image data for the pixel of interest and the detected dot formation presence / absence;
Have
The dot formation determination unit is a unit that makes it easier to determine that there is dot formation as the gradation value of the pixel of interest increases.
The converted unit selection means has a correspondence relationship with the total value that is larger than the first predetermined value and smaller than the second predetermined value for a total value smaller than the first predetermined value. The gist is that the unit is stored in association with a unit far from the pixel of interest rather than the total value.
[0009]
The image processing method of the present invention corresponding to the above image processing apparatus is
An image processing method for converting image data expressed by gradation values for each pixel constituting an image into dot data indicating the presence or absence of dot formation for each pixel,
A first step of grouping a predetermined plurality of pixels adjacent to each other as a unit, and converting the image data into the dot data in units;
A second step of storing, with respect to at least some of the pixels constituting the converted unit, the presence or absence of dot formation in a state in which the pixel positions in the unit can be identified;
With
The first step includes
A sub-step (1-a) of calculating a total sum of gradation values for each pixel constituting the unit and selecting a predetermined number of units from the converted units according to a correspondence relationship with the total value; ,
A sub-step (1-b) for sequentially setting a pixel of interest for determining whether or not to form dots in a unit to be converted into dot data;
A sub-step (1-c) for detecting, with respect to the selected predetermined number of units, whether or not dots are formed at a pixel at a predetermined position with respect to the pixel position of the target pixel;
A sub-step (1-d) for determining the presence / absence of dot formation for the target pixel based on the gradation value of the image data for the target pixel and the detected dot formation presence / absence;
Have
The sub-step (1-d) is a step that makes it easier to determine that there is dot formation as the gradation value of the pixel of interest increases.
In the substep (1-a), as a correspondence relationship with the total value, the total value smaller than the first predetermined value is larger than the first predetermined value and larger than the second predetermined value. The summary is that it is a step of storing a unit far from the pixel of interest in association with it, rather than a smaller sum value.
[0010]
In such an image processing apparatus and image processing method, a predetermined number of pixels are grouped as a unit, and image data is converted into dot data in units. At this time, the dot formation status in the already converted unit is detected, and the presence or absence of dot formation is determined in consideration of the detection result. The dot formation status in the converted unit is detected as follows. First, for at least some of the pixels constituting the converted unit, the presence / absence of dot formation is stored in a state where the pixel position in the unit can be identified. Then, the sum of gradation values for each pixel constituting the unit to be converted into dot data is calculated, and a predetermined number of units are selected from the already converted units according to the correspondence with the sum value. . Here, as a correspondence relationship with the total value, the total value smaller than the first predetermined value is larger than the total value smaller than the first predetermined value and smaller than the second predetermined value. A relationship that associates units far from the target pixel is set. Further, the number of units to be selected according to the correspondence relationship with the total value may be one or plural. Further, the pixel position for detecting the presence or absence of dot formation in the selected unit may be the same pixel position as the target pixel in the unit, or a different pixel position depending on the pixel position of the target pixel. It is also possible. Next, in the unit to be converted, the target pixel for determining the presence / absence of dot formation is sequentially set, and the presence / absence of dot formation at a pixel at a predetermined position with respect to the pixel position of the target pixel is selected. Detect for a number of units.
[0011]
In this way, if the dot formation status in the already converted unit is detected and the presence or absence of dot formation is determined in consideration of the detection result, even if the image data is converted into dot data on a unit basis, dots are not detected in the unit. It can avoid that the position formed becomes a fixed pattern over a plurality of units. Accordingly, it is possible to reliably prevent the periodic pattern from being visually recognized and the image quality from deteriorating. At this time, when dots are formed sparsely, it is considered that the period of a constant pattern generated over a plurality of units becomes longer. Therefore, when the sum of the gradation values becomes a small value, when selecting from the target pixel to a unit far from the target pixel, detecting the dot formation status for these selected units, and determining whether or not the target pixel is formed. If this is taken into consideration, it is possible to effectively suppress the formation of dots in a periodic pattern.
[0012]
Here, only the unit far from the target pixel may be associated with the total value smaller than the first predetermined value, but it is larger than the first predetermined value and larger than the second predetermined value. A unit that is relatively close to the target pixel and that is associated with a smaller total sum value may also be associated. This is preferable because not only the formation of a pattern having a long period but also the formation of a pattern having a relatively short period can be effectively avoided.
[0013]
In such an image processing apparatus, not only for the total value smaller than the first predetermined value, but also for the total value larger than the second predetermined value, the second predetermined value is greater than the first predetermined value. Rather than the sum value smaller than the value, the unit from the target pixel to a distant unit may be associated. At this time, a farther unit may be associated with the sum value smaller than the first predetermined value and stored in association with the sum value larger than the second predetermined value.
[0014]
Similar to the case where dots are formed sparsely, when the dots are formed at a high density and pixels where dots are not formed are generated sparsely, the period of the pattern generated over a plurality of units is considered to be long. Therefore, even when the sum of the gradation values is large, as in the case where the sum is small, select from the pixel of interest to a distant unit, detect the dot formation status for these selected units, and If this is taken into consideration when determining the presence or absence of dot formation for a pixel, it is possible to effectively suppress the formation of dots in a periodic pattern. However, it is considered that there are the following differences between the case where dots are formed sparsely and the case where dots are formed with high density.
[0015]
When dots are formed at a high density, the dots may be formed larger due to the interaction between adjacent dots. In addition, when forming dots by ejecting ink droplets on a print medium, the dots are formed larger due to the effect of ink bleeding, so if dots are formed at a high density, the adjacent dots partially overlap each other. May end up. In such a case, even if pixels in which dots are not formed are sparsely formed in a fixed pattern, the surrounding pattern is formed to be large, so that the periodic pattern is not noticeable. Therefore, in areas where the gradation value is large and the dots are formed at a high density, the area where the gradation value is small and the dots are formed sparsely does not consider whether or not dots are formed in a distant unit. In both cases, the image quality is not expected to deteriorate. Therefore, even when the sum total value of the gradation values becomes a large value, as in the case where the sum total value of the gradation values is small, as far as the unit is selected, the small sum value is larger than the large sum value. However, if even farther units are selected, deterioration of image quality can be effectively suppressed.
[0016]
Further, in such an image processing device, when a dot is detected at a pixel at a predetermined position with respect to the pixel of interest when the sum of the gradation values is small, it is difficult to form a dot for the pixel of interest. Also good. Here, the case where a dot is detected at a pixel at a predetermined position may be a case where a dot is detected at any of a plurality of pixels at a predetermined position, or a dot is detected at all of the plurality of pixels. It can also be the case.
[0017]
In the region where the total sum of the gradation values is small, dots are formed sparsely. Therefore, when dots are formed in pixels at a predetermined position in this way, it is possible to prevent dots from being formed in a periodic pattern over a plurality of units if it is difficult to form dots in the pixel of interest. can do. As a result, even when the image data is converted into dot data in units, it is possible to avoid deterioration of the image quality due to the periodic pattern being visually recognized.
[0018]
In particular, in a region where dots are sparsely formed, it is considered rare that dots are formed in all of a plurality of pixels. Therefore, when dots are formed in any of the predetermined pixels, the generation of a periodic pattern can be effectively suppressed by suppressing the formation of dots at the target pixel. Further, in a region where dots are formed at a certain density, it is considered that there is a high possibility that dots are formed in any of the plurality of pixels. Therefore, when all dots are formed in a predetermined pixel, it is possible to surely avoid the formation of dots with a periodic pattern by suppressing the formation of dots at the target pixel. This is preferable.
[0019]
Furthermore, in such an image processing apparatus, in a region where the total sum of gradation values is small, when dots are formed at pixels at predetermined positions, dots may not be formed at the target pixel.
[0020]
This is preferable because even when image data is converted into dot data in units, it is possible to easily avoid the formation of dots in a fixed pattern.
[0021]
In addition, in such an image processing device, when the sum of the gradation values is large, if a dot is not detected in a pixel at a predetermined position with respect to the target pixel, the dot may be easily formed for the target pixel. good. Here, the case where a dot is not detected at a pixel at a predetermined position may be a case where a dot is not detected at any of a plurality of pixels at a predetermined position, or a dot is not detected at any of the plurality of pixels. It can also be a case.
[0022]
In a region where the total sum of gradation values is large, pixels in which dots are not formed are sparsely generated. That is, when image data is converted into dot data in units of units, pixels in which dots are not formed may occur in a certain pattern and deteriorate image quality. Therefore, if dots are not formed in the pixel at the predetermined position, it is possible to prevent the formation of a pixel in which a dot is not formed across a plurality of units in a periodic pattern by making it easy to form a dot in the target pixel. can do.
[0023]
In particular, in a region where pixels where dots are not formed hardly occur, it is rare that dots are not formed at all of the pixels at a predetermined position. Therefore, when dots are not formed in any pixel at a predetermined position, it is possible to effectively suppress the generation of a periodic pattern by making it easier to form dots in the pixel of interest. Become. In a region where pixels where dots are not formed are generated at a certain density, it is considered that there is a high possibility that dots are not formed in any of the plurality of pixels. Therefore, when dots are not formed in all pixels at a predetermined position, it is preferable to make it easier to form dots in the pixel of interest because it is possible to reliably avoid the generation of a periodic pattern. .
[0024]
Further, in such an image processing apparatus, in a region where the sum total value of the gradation values is large, when a dot is not formed on a pixel at a predetermined position, a dot may be formed on the pixel of interest.
[0025]
This is preferable because even when image data is converted into dot data in units, it is possible to easily avoid occurrence of pixels in which dots are not formed in a fixed pattern.
[0026]
In the above-described image processing apparatus, the presence / absence of dot formation for each pixel constituting the unit is determined using a so-called error diffusion method, and the threshold value of the error diffusion method referred to in such determination is converted. It is good also as changing according to the dot formation presence or absence in the unit.
[0027]
If the threshold value of the error diffusion method is changed, the detection result of dot formation presence / absence at a predetermined pixel position in the converted unit can be easily reflected in determination of dot formation presence / absence in the target pixel. In addition, the error diffusion method quickly eliminates the tone error by diffusing the tone error generated in the pixel of interest to the surrounding undetermined pixels and reflecting it in the determination of dot formation in these surrounding pixels. It has a function to make it. From this, if the presence / absence of dot formation for the pixel of interest is determined using the error diffusion method, a large gradation error will occur in the pixel of interest by reflecting the detection result of dot formation in the converted unit. Even if it occurs, it is preferable because this can be resolved quickly and a higher quality image can be expressed.
[0028]
In such an image processing apparatus that determines the presence / absence of dot formation using the error diffusion method, when dot formation is detected at a pixel at the predetermined position with respect to the target pixel, It may be determined that dots are not formed.
[0029]
If dot formation is detected at a pixel at a predetermined position, if it is determined that no dot is to be formed on the pixel of interest, even if image data is converted to dot data in units, It is preferable because it is possible to easily avoid the formation of dots in the pattern. In addition, since the presence / absence of dot formation is determined using the error diffusion method, even if a large gradation error occurs in the pixel of interest by determining that dots are not formed, this error can be quickly eliminated. This is possible and there is no risk of image quality deterioration.
[0030]
The above-described image processing apparatus can be suitably applied to a printing apparatus that forms dots on a print medium and prints an image. That is, the image processing apparatus of the present invention can convert image data to dot data quickly by converting in units. In addition, by detecting the dot formation status at a predetermined pixel position in the converted unit, it is possible to avoid the formation of dots in a constant pattern. Therefore, it is preferable to apply such an image processing apparatus to a printing apparatus, because it is possible to avoid deterioration of image quality while shortening the time required for printing by quickly converting image data into dot data.
[0031]
The present invention can also be realized using a computer by causing a computer to read a program that realizes the above-described image processing method. Therefore, the present invention includes the following program or a mode as a recording medium on which the program is recorded. That is, the program of the present invention corresponding to the image processing method described above is
A program for realizing, using a computer, a method for converting image data expressed by gradation values for each pixel constituting an image into dot data indicating the presence or absence of dot formation for each pixel. ,
A first function of grouping a plurality of adjacent pixels as a unit, and converting the image data into the dot data in units;
A second function for storing the presence / absence of dot formation in a state in which the pixel position in the unit can be identified for at least some of the pixels constituting the converted unit;
As well as
The first function further includes:
A function (1-a) for calculating a sum of gradation values for each pixel constituting the unit, and selecting a predetermined number of units from the converted units according to a correspondence relationship with the sum value;
A function (1-b) for sequentially setting a pixel of interest for determining the presence or absence of dot formation in the unit to be converted into the dot data;
A function (1-c) for detecting the presence or absence of dot formation at a pixel at a predetermined position with respect to the pixel position of the target pixel for the selected predetermined number of units;
A function (1-d) for determining the presence / absence of dot formation for the pixel of interest based on the gradation value of the image data for the pixel of interest and the detected dot formation presence / absence;
Have
The function (1-d) is a function that makes it easier to determine that there is dot formation as the gradation value of the pixel of interest increases.
The function (1-a) is greater than the first predetermined value and larger than the second predetermined value for the total value smaller than the first predetermined value as a correspondence relationship with the total value. The gist of the present invention is that it is a function that stores a unit far from the pixel of interest in association with it rather than a small sum value.
[0032]
The recording medium of the present invention corresponding to the above-described image processing method is
A computer-readable recording program for realizing a method of converting image data expressed by gradation values for each pixel constituting an image into dot data indicating the presence or absence of dot formation for each pixel is recorded. Recording medium,
A first function of grouping a plurality of adjacent pixels as a unit, and converting the image data into the dot data in units;
A second function for storing the presence / absence of dot formation in a state in which the pixel position in the unit can be identified for at least some of the pixels constituting the converted unit;
Realized,
The first function further includes:
A function (1-a) for calculating a sum of gradation values for each pixel constituting the unit, and selecting a predetermined number of units from the converted units according to a correspondence relationship with the sum value;
A function (1-b) for sequentially setting a pixel of interest for determining the presence or absence of dot formation in the unit to be converted into the dot data;
A function (1-c) for detecting the presence or absence of dot formation at a pixel at a predetermined position with respect to the pixel position of the target pixel for the selected predetermined number of units;
A function (1-d) for determining the presence / absence of dot formation for the pixel of interest based on the gradation value of the image data for the pixel of interest and the detected dot formation presence / absence;
Is realized using a computer,
The function (1-d) is a function that makes it easier to determine that there is dot formation as the gradation value of the pixel of interest increases.
The function (1-a) is greater than the first predetermined value and larger than the second predetermined value for the total value smaller than the first predetermined value as a correspondence relationship with the total value. The gist of the present invention is that it is a function that stores a unit far from the pixel of interest in association with it rather than a small sum value.
[0033]
If such a program or a program stored in a recording medium is read into a computer and the above-described various functions are executed by the computer, the image data is converted into dot data in units and quickly converted. By detecting the dot formation status in the converted unit, it is possible to avoid the formation of dots in a fixed pattern, so there is no risk of image quality deterioration.
[0034]
DETAILED DESCRIPTION OF THE INVENTION
In order to more clearly describe the operation and effect of the present invention, embodiments of the present invention will be described below in the following order.
A. Embodiment:
B. Device configuration:
C. Overview of image data conversion process:
D. First embodiment:
D-1. Tone number conversion processing performed in units:
D-2. In-unit multi-value processing:
D-3. In-unit multilevel processing of the first embodiment:
D-4. Variation:
E. Second embodiment:
E-1. In-unit multi-value processing of the second embodiment:
E-2. Modification of the second embodiment:
[0035]
A. Summary of the invention:
Before starting the detailed description of the embodiments, the outline of the present invention will be briefly described for convenience of understanding. FIG. 1 is an explanatory diagram showing an outline of the present invention, taking a printing system as an example. The printing system shown in FIG. 1 includes a computer 10 as an image processing apparatus, a color printer 20 and the like. When the computer 10 receives color image data expressed by gradation values of RGB colors from an image device such as a digital camera or a color scanner, the computer 10 expresses the image data by the presence or absence of formation of each color dot that can be printed by the color printer 20. Is converted to print data. Such conversion of image data is performed using a dedicated program called the printer driver 12. Note that color image data for each color of RGB can also be created by the computer 10 using various application programs.
[0036]
The printer driver 12 includes a plurality of modules such as a resolution conversion module, a color conversion module, a gradation number conversion module 14, and an interlace module. Processing performed in each of these modules will be described later. Among these, the gradation number conversion module 14 performs a process of converting the image data into data expressed by the presence or absence of dot formation. Further, in the gradation number conversion module of this embodiment, in order to speed up the processing, the unit unit processing is performed as follows. That is, a unit is generated by grouping a plurality of adjacent pixels, and whether or not dots are formed for each pixel is collectively determined for each unit. A method for determining whether or not to form dots in units will be described later.
[0037]
In FIG. 1, a conceptual diagram indicated by an arrow below a frame representing the gradation number conversion module 14 shows a state in which image data is converted into dot data in the gradation number conversion module 14. . The small squares indicated by broken lines in the conceptual diagram schematically show the pixels. These four pixels constitute one unit. The units shown with fine diagonal lines in the figure indicate units that are being processed, that is, units that are converting image data into dot data. A region indicated by a rough oblique line indicates a region converted to dot data after the presence / absence of dot formation is determined. When the determination of dot formation is performed in units, the same processing is performed for each unit. For this reason, particularly when an area where the gradation value of the image data does not change so much is processed, there is a possibility that dots are formed in a constant pattern across a plurality of units. If a region in which dots are formed in a certain pattern is generated, there is a concern that this is visually recognized and the image quality is deteriorated.
[0038]
The conceptual diagram in FIG. 1 shows a case where dots are formed at the same pixel position in a continuous unit as an example in which dots are formed in a constant pattern. As shown in the drawing, on the left side of the unit being processed (the unit with fine slashes), two units in which dots are formed in the pixel in the upper left corner are continuous. If a dot is formed in the pixel in the upper left corner even in the unit being processed, dots will be formed in the pixel at the same position in three units in succession, and this will be visually recognized as a pattern and image quality will deteriorate. There is a fear. Of course, an example in which dots are formed in a fixed pattern is not limited to the case where dots are formed at the same pixel position in successive units, but where dots are formed in a fixed pattern with multiple units as a period. Can also happen.
[0039]
In order to avoid the occurrence of such a problem, the gradation number conversion module 14 in the printing system shown in FIG. 1 stores in the gradation number conversion result storage unit pixels that have been determined to form dots for units that have already been processed. Remember. When determining the presence or absence of dot formation on a unit basis, the determination result is read from the gradation number conversion result storage unit, and determination is made while avoiding the formation of dots in a fixed pattern. For example, in the example shown in the conceptual diagram in FIG. 1, dots are formed at the same pixel position in consecutive units, so that dots are not continuously formed at pixels at the same position. It is determined whether or not dots are formed. In this manner, the presence / absence of dot formation is determined while avoiding the formation of dots in a pattern that is easily visible. In this way, it is possible to print a high-quality image without deteriorating the image quality while performing quick processing by determining the presence or absence of dot formation in units.
[0040]
Also, for example, in a region where dots are formed sparsely, such as being formed at a rate of one in several units, the period in which dots are formed in a constant pattern is expected to be longer. Similarly, even in a region where dots are formed in most pixels, it is conceivable that pixels in which dots are not formed occur in a constant pattern across a plurality of units. Therefore, in these areas, it is possible to more effectively suppress the deterioration of the image quality if the dot formation state is taken into consideration for farther units. Hereinafter, such image processing will be described in detail based on examples.
[0041]
B. Device configuration:
FIG. 2 is an explanatory diagram illustrating a configuration of a computer 100 as the image processing apparatus according to the present exemplary embodiment. The computer 100 is a well-known computer configured by connecting a ROM 104, a RAM 106, and the like with a bus 116 around a CPU 102.
[0042]
Connected to the computer 100 are a disk controller DDC 109 for reading data on the flexible disk 124 and the compact disk 126, a peripheral device interface PIF 108 for exchanging data with peripheral devices, a video interface VIF 112 for driving a CRT 114, and the like. Has been. A color printer 200, a hard disk 118, and the like, which will be described later, are connected to the PIF 108. Further, if the digital camera 120, the color scanner 122, and the like are connected to the PIF 108, it is possible to print an image captured by the digital camera 120 or the color scanner 122. If the network interface card NIC 110 is attached, the computer 100 can be connected to the communication line 300 to acquire data stored in the storage device 310 connected to the communication line.
[0043]
FIG. 3 is an explanatory diagram showing a schematic configuration of the color printer 200 of the first embodiment. The color printer 200 is an ink jet printer capable of forming dots of four color inks of cyan, magenta, yellow, and black. Of course, in addition to these four color inks, an ink jet printer capable of forming ink dots of a total of six colors including cyan (light cyan) ink having a low dye density and magenta (light magenta) ink having a low dye density is used. You can also. In the following, cyan ink, magenta ink, yellow ink, black ink, light cyan ink, and light magenta ink are abbreviated as C ink, M ink, Y ink, K ink, LC ink, and LM ink, respectively. There shall be.
[0044]
As shown in the figure, the color printer 200 has a mechanism for driving the print head 241 mounted on the carriage 240 to eject ink and forming dots, and the carriage 240 is reciprocated in the axial direction of the platen 236 by the carriage motor 230. A mechanism for moving the printing paper P by a paper feed motor 235, and a control circuit 260 for controlling dot formation, carriage 240 movement, and printing paper conveyance.
[0045]
An ink cartridge 242 that stores K ink and an ink cartridge 243 that stores various inks of C ink, M ink, and Y ink are mounted on the carriage 240. When the ink cartridges 242 and 243 are mounted on the carriage 240, each ink in the cartridge is supplied to ink discharge heads 244 to 247 for each color provided on the lower surface of the print head 241 through an introduction pipe (not shown). Each of the ink discharge heads 244 to 247 for each color is provided with one set of nozzle rows in which 48 nozzles Nz are arranged at a constant nozzle pitch k.
[0046]
The control circuit 260 includes a CPU 261, a ROM 262, a RAM 263, and the like. The control circuit 260 controls main scanning and sub-scanning of the carriage 240 by appropriately driving the carriage motor 230 and the paper feed motor 235, and controls each nozzle based on print data supplied from the computer 100. Ink droplets are discharged at an appropriate timing. Thus, the color printer 200 can print a color image by forming ink dots of respective colors at appropriate positions on the print medium under the control of the control circuit 260.
[0047]
Various methods can be applied to the method of ejecting ink droplets from the ink ejection heads of the respective colors. That is, a method of ejecting ink using a piezoelectric element, a method of ejecting ink droplets by generating bubbles in the ink passage with a heater arranged in the ink passage, and the like can be used. Also, instead of ejecting ink, use a method that uses ink transfer to form ink dots on printing paper using a phenomenon such as thermal transfer, or a method that uses static electricity to attach toner powder of each color onto the print medium. It is also possible to do.
[0048]
Furthermore, the size of the ink dots formed on the printing paper can be controlled by controlling the size of the ejected ink droplets, or by ejecting a plurality of ink droplets at a time and controlling the number of ejected ink droplets. A so-called variable dot printer capable of controlling the height can also be used.
[0049]
The color printer 200 having the hardware configuration as described above drives the carriage motor 230 to move the ink ejection heads 244 to 247 for each color in the main scanning direction with respect to the printing paper P, and the paper feed motor. By driving 235, the printing paper P is moved in the sub-scanning direction. The control circuit 260 drives the nozzles at appropriate timing to discharge ink droplets while repeating main scanning and sub-scanning of the carriage 240 according to the print data, so that the color printer 200 prints a color image on the printing paper. is doing.
[0050]
C. Overview of image data conversion process:
FIG. 4 is a flowchart showing a flow of processing in which the computer 100 as the image processing apparatus of the present embodiment converts image data into print data by applying predetermined image processing to the received image data. Such processing is started when the operating system of the computer 100 activates the printer driver 12. Hereinafter, the image data conversion processing of the present embodiment will be briefly described with reference to FIG.
[0051]
When the image data conversion process is started, the printer driver 12 first starts reading RGB color image data to be converted (step S100). Next, the resolution of the captured image data is converted to a resolution for printing by the color printer 200 (step S102). When the resolution of color image data is lower than the print resolution, new data is generated between adjacent image data by performing linear interpolation. Conversely, when the resolution is higher than the print resolution, the data is thinned out at a fixed rate. Thus, the resolution of the image data is converted into the printing resolution.
[0052]
When the resolution is converted in this way, color conversion processing of color image data is performed (step S104). Color conversion processing refers to color image data expressed by a combination of R, G, and B gradation values, by a combination of gradation values of each color used in the color printer 200 such as C, M, Y, and K. This is a process of converting into expressed image data. The color conversion process can be quickly performed by referring to a three-dimensional numerical table called a color conversion table.
[0053]
The printer driver 12 starts the tone number conversion process following the color conversion process (step S106). The gradation number conversion process is the following process. By the color conversion process, the RGB image data is converted into gradation data of C, M, Y, and K colors. The gradation data of each color is data having 256 gradations with gradation values 0 to 255. On the other hand, the color printer 200 of the present embodiment can only take one of the states of “form dots” or “do not form dots”. Therefore, it is necessary to convert the gradation data of each color having 256 gradations into image data expressed in 2 gradations that can be expressed by the color printer 200. A process for converting the number of gradations is a gradation number conversion process. As will be described later, the printer driver 12 according to the present embodiment enables rapid processing by collecting a predetermined number of pixels into units and performing gradation number conversion processing on a unit basis. For the processed unit, the pixel determined to form a dot is stored, and the number of gradations is converted while reflecting this data. By doing this, it is possible to reliably avoid the occurrence of a phenomenon in which dots are formed in a constant pattern, which is a concern when the gradation number conversion processing is performed on a unit basis. The tone number conversion process will be described in detail later.
[0054]
When the tone number conversion process is thus completed, the printer driver starts the interlace process (step S108). The interlacing process is a process of rearranging image data (dot data) converted into a format representing the presence / absence of dot formation into an order to be transferred to the color printer 200 in consideration of the dot formation order. The printer driver 12 outputs the image data finally obtained by performing the interlace processing to the color printer 200 as print data (step S110). The color printer 200 forms ink dots of each color on the print medium according to the print data. As a result, a color image corresponding to the image data is printed on the print medium.
[0055]
D. First embodiment:
Hereinafter, the tone number conversion process of the first embodiment performed in the image data conversion process shown in FIG. 4 will be described. As described above, in the gradation number conversion process of this embodiment, a plurality of pixels are grouped into units, and the gradation number conversion process is performed in units. Therefore, as a preparation for explaining the gradation number conversion process of the first embodiment, first, a method of performing the gradation number conversion process in units will be described, and then, in the gradation number conversion process of the first embodiment. A process of reflecting the determination result of the processed unit in the determination of the presence / absence of dot formation will be described.
[0056]
D-1. Tone number conversion processing performed in units:
FIG. 5 is a flowchart showing the flow of gradation number conversion processing performed in units. This process is performed by the CPU 102 of the computer 100. Note that the color printer 200 of this embodiment is a printer that can form ink dots of four colors C, M, Y, and K as described above, and the processing shown in FIG. 5 is also performed for each color. However, in the following description, in order to avoid complicated explanation, the explanation will be made without specifying the color. Unless otherwise specified, when the explanation is made without specifying the color, the same processing is performed for each color. It means to be done.
[0057]
When the unit-unit gradation number conversion process is started, first, the position of the unit to be processed is set (step S200). In the gradation number conversion process performed in units, a predetermined plurality of adjacent pixels are grouped as a unit, and it is determined in units of whether or not to form dots in each pixel. Therefore, when the unit-unit gradation number conversion process is started, first, a process of setting the position of a unit (processing unit) for which it is determined whether or not dots are formed in the image is performed.
[0058]
FIG. 6 is an explanatory diagram conceptually showing a state in which a predetermined plurality of adjacent pixels are grouped to set the position of the processing unit. In FIG. 6, a plurality of small squares displayed schematically represent pixels. As shown in the figure, the image is composed of a plurality of pixels arranged in a grid pattern. In the example shown in FIG. 6A, a processing unit is set by grouping four adjacent pixels arranged in two vertical and horizontal columns. In FIG. 6A, a thick broken line surrounding four pixels schematically shows that these pixels are set as a processing unit. For convenience of explanation, among the four pixels constituting the unit, the upper left pixel is called “Pa”, the upper right pixel is “Pb”, the lower left pixel is “Pc”, and the lower right pixel is called “Pd”. A distinction shall be made. In the following description, the unit is assumed to be composed of four pixels arranged in two rows and columns, but of course, the unit is not limited to such a configuration. For example, the processing unit may be set by grouping four adjacent pixels as shown in FIG. Further, the number of pixels grouped as a processing unit is not limited to four. For example, two pixels may be set as a processing unit as shown in FIGS. 6C and 6D.
[0059]
After the processing unit is set, image data and diffusion error are read for each pixel in the unit (step S202 in FIG. 5). The image data read here is gradation data of each color of C, M, Y, and K obtained by color conversion by the above-described color conversion process (step S104 in FIG. 4). The diffusion error is an error in which the gradation error caused by converting the number of gradations is diffused from the peripheral unit to each pixel of the processing unit. Since the image data is gradation data for each color of C, M, Y, and K, a gradation error occurs for each of these colors, and therefore the diffusion error is also gradation data for each of these colors. As will be described later, a gradation error is generated along with the conversion of the number of gradations, and the generated error is diffused as a diffusion error to surrounding pixels.
[0060]
Here, as shown in FIG. 6A, the processing unit is assumed to be composed of four pixels, pixel Pa, pixel Pb, pixel Pc, and pixel Pd. Therefore, in the process of step S202 in FIG. For these pixels, image data DTa, DTb, DTc, DTd and diffusion errors EDa, EDb, EDc, EDd are read. These data read for the processing unit are stored in the RAM 106 of the computer 100.
[0061]
When the image data and the diffusion error of each pixel are read in this way, a process for determining whether or not dots are formed is started for each pixel constituting the processing unit (step S204). Here, as shown in FIG. 6A, the processing unit is assumed to be composed of four pixels. Therefore, in the process of step S204, the presence or absence of dot formation is determined for these four pixels in a predetermined order. Go. In the present specification, the process for determining the presence or absence of dot formation for each pixel in the unit is sometimes referred to as “in-unit multi-value processing”. Details of the multi-value processing in the unit will be described later.
[0062]
When the determination of the presence / absence of dot formation for each pixel constituting the processing unit is completed, a process of diffusing the gradation error generated in the processing unit to the peripheral units with the determination is performed (step S206). The gradation error generated in the processing unit is the gradation value of the image data to be expressed in the entire processing unit and the gradation value actually expressed by forming dots at each pixel in the unit. It is an error representing a deviation. A method for calculating the gradation error occurring in the processing unit will be described in the explanation of the multi-value processing in the unit. In step S206, a process of diffusing the gradation value indicating the deviation to surrounding unprocessed units as a gradation error generated in the processing unit is performed.
[0063]
FIG. 7 is an explanatory diagram conceptually showing a state in which the gradation error generated in the processing unit is diffused to surrounding unprocessed units. Each of the plurality of small squares shown in FIG. 7 schematically represents a pixel. A large square indicated by a thick broken line indicates a processing unit, and a large square indicated by a thin broken line indicates an unprocessed unit adjacent to the processing unit. The gradation error generated in the processing unit is diffused at a predetermined ratio to the pixels constituting the surrounding unprocessed units. In FIG. 7, the black arrows extending from the processing unit to the surrounding six pixels schematically indicate that the gradation error generated in the processing unit is diffused to these six pixels at a predetermined ratio. Yes.
[0064]
FIG. 8 illustrates a state in which the ratio of diffusing gradation errors generated in the processing unit to pixels constituting unprocessed units around the processing unit is set. In FIG. 8, a large square indicated by a thick solid line represents a processing unit, and a large square indicated by a thin broken line adjacent to the processing unit represents an unprocessed unit around the processing unit. is there. A plurality of small squares indicated by thin solid lines in these unprocessed units represent pixels in which an error from the processing unit is diffused.
[0065]
In the example shown in FIG. 8A, the gradation error generated in the processing unit is diffused to the two pixels on the right side of the processing unit at a ratio of 1/8, and 2 pixels on the lower side of the processing unit. Similarly, the gradation error is diffused to each pixel at a rate of 1/8. Further, the gradation error is diffused at a ratio of 1/4 to the lower left pixel or the lower right pixel of the processing unit. If the gradation error is diffused at such a ratio, the gradation error is equally diffused to the peripheral units of the processing unit at a ratio of 1/4 of the gradation error. In the example shown in FIG. 8B, the gradation error generated in the processing unit is diffused at the same rate to all the pixels constituting the unit around the processing unit. Of course, errors that diffuse to the same unit may be diffused together in one pixel of that unit.
[0066]
In step S206 of FIG. 5, the gradation error generated in the processing unit is diffused to the peripheral units as described above. When all the processes described above are completed for the processing unit, it is determined whether or not the process has been completed for all the units constituting the image (step S208). If there are any unprocessed units, the process returns to step S200. The following series of processes is repeated. Thus, when the processing for all the units is completed, the gradation number conversion processing for each unit is completed.
[0067]
When the gradation number conversion process is performed on a unit basis as described above, the following advantages can be obtained as compared with the case of processing for each pixel. That is, when the gradation number conversion process is performed for each pixel, the diffusion error stored in association with the pixel must be read every time it is determined whether or not a dot is formed for one pixel. It is necessary to diffuse the error caused by the determination of dot formation to the surrounding undetermined pixels. For this reason, when the gradation number conversion process is performed for each pixel, the processing time tends to be long. In contrast, when the gradation number conversion process is performed in units, it is possible to read the image data in units and diffuse the gradation errors, and then write the diffusion errors into the memory in units. If data reading and writing are performed in unit units in this way, the time required for processing can be shortened.
[0068]
D-2. In-unit multi-value processing:
The in-unit multilevel conversion process (step S204 in FIG. 5) performed in the unit-unit gradation number conversion process described above will be described. FIG. 9 is a flowchart showing the flow of the in-unit multilevel processing. Hereinafter, it demonstrates according to a flowchart.
[0069]
When the intra-unit multilevel processing is started, first, one pixel (a target pixel) for determining the presence or absence of dot formation is set from the pixels constituting the processing unit (step S300). Since the unit is composed of a plurality of pixels as described above, one pixel of interest is selected by determining whether or not dots are formed pixel by pixel in a predetermined order for the plurality of pixels. Here, the four pixels constituting the unit shown in FIG. 6A are subjected to dot formation presence / absence in the order of pixel Pa, pixel Pb, pixel Pc, pixel Pd from the upper left pixel to the lower right pixel. In step S300, first, the pixel Pa at the upper left corner is set as the target pixel.
[0070]
Next, correction data Cx for the set pixel of interest (here, pixel Pa) is calculated (step S302). The correction data Cx can be obtained by adding the image data of the target pixel and the diffusion error diffused and stored in the target pixel. Since the image data and the diffusion error for each pixel constituting the processing unit have already been read in step S202 during the gradation number conversion processing shown in FIG. Adds the image data of the pixel of interest and the diffusion error from these data to calculate the correction data Cx.
[0071]
After the correction data Cx for the pixel of interest Pa is calculated, the magnitude relationship between the obtained correction data Cx and the predetermined threshold th is determined (step S304). If the correction data Cx is larger, it is determined that a dot is to be formed on the target pixel (here, pixel Pa) (step S3306). Otherwise, it is determined that no dot is formed on the target pixel (step S308). . The result of the determination is stored in a variable indicating the determination result for each pixel.
[0072]
When the presence / absence of dot formation for the target pixel is determined in this way, the gradation error generated in the target pixel is calculated in accordance with the determination (step S310). The gradation error is a gradation value expressed in the pixel of interest by forming a dot or not forming a dot (hereinafter, this gradation value is referred to as a result value). It can be calculated by subtracting from Cx.
[0073]
After the gradation error generated in the target pixel is calculated, it is determined whether or not the processing of all the pixels in the processing unit has been completed (step S312). If there are unprocessed pixels remaining (step S312: no). Then, a process of diffusing an error to pixels around the target pixel among these unprocessed pixels is performed (step S314). This process will be described with reference to FIG.
[0074]
FIG. 10 is an explanatory diagram schematically showing a state in which the presence / absence of dot formation is determined on a unit basis by selecting each pixel of interest from the pixels in the processing unit and determining the presence / absence of dot formation. It is. Four squares in the figure indicate the pixel Pa, the pixel Pb, the pixel Pc, and the pixel Pd, respectively, constituting the processing unit. DTa and Eda shown in the square indicating the pixel Pa conceptually show that the image data DTa and the diffusion error Eda are read into the RAM 106 in association with the pixel Pa. Similarly, image data DTb and diffusion error EDb are associated with pixel Pb, image data DTc and diffusion error EDc are associated with pixel Pc, and image data DTd and diffusion error EDd are associated with pixel Pd, respectively. Stored in the RAM 106.
[0075]
As described above, the target pixel is initially set to the pixel Pa among these four pixels. The correction data Cx is calculated by reading out the image data and the diffusion error for the pixel Pa, and the presence or absence of dot formation for the pixel Pa is determined by comparing with the threshold value th. At the time when the presence or absence of dot formation is determined for the pixel Pa, as shown in FIG. 10A, three pixels of the pixel Pb, the pixel Pc, and the pixel Pd remain as undetermined pixels in the same unit. Therefore, the gradation error generated in the pixel Pa is equally distributed to each of these three pixels by 1/3, and the error is added to the diffusion error stored for each pixel. For example, as shown in FIG. 10A, the diffusion error EDb is already stored in association with the pixel Pb. The gradation error generated in the pixel Pa is distributed to the pixel Pb by 1/3 of the gradation value and added to the diffusion error EDb. As a result, the diffusion error EDb stored in association with the pixel Pb is updated to a new diffusion error EDb to which the error from the pixel Pa is added.
[0076]
Similar processing is performed for the other pixels Pc and Pd. Briefly, when the error from the pixel Pa is diffused to the pixel Pc, the diffusion error EDc stored in the pixel Pc and the error from the pixel Pa are added to correspond to the pixel Pc as a new diffusion error EDc. It is memorized. Similarly, for the pixel Pd, the diffusion error EDd stored in the pixel Pd and the error from the pixel Pa are added and stored as a new diffusion error EDd in association with the pixel Pd. In FIG. 10A, the white arrow displayed from the pixel Pa toward the other three pixels schematically indicates that the error generated in the pixel Pa is diffused to these three pixels. Is. Note that the gradation error does not necessarily need to be distributed evenly to the surrounding undetermined pixels, and can be distributed to each pixel at a predetermined ratio.
[0077]
After the gradation error is diffused to the peripheral pixels in the unit as described above, the process returns to step S300 in FIG. 9 to perform a process for setting a new pixel of interest. Since the pixel Pa in the processing unit has already been determined for dot formation, in step S300, the pixel Pb right next to the pixel Pa is set as the pixel of interest. For the new pixel of interest set in this way, processing similar to that described above is performed. Hereinafter, the difference from the processing in the pixel Pa will be briefly described.
[0078]
When the process for the pixel Pb is performed, the correction data Cx is first calculated as in the case of the pixel Pa (step S302). Here, the gradation error previously generated in the pixel Pa is diffused into the diffusion error EDb stored in association with the pixel Pb, and updated as a new diffusion error EDb. Therefore, the correction data Cx for the pixel Pb is calculated by adding the image data DTb for the pixel Pb and the new diffusion error EDb in which the gradation error from the pixel Pa is diffused. The magnitude relationship between the correction data Cx thus calculated and the predetermined threshold th is determined (step S304). If the correction data Cx is larger, it is determined that a dot is to be formed in the pixel Pb (step S306), otherwise. It is determined that no dot is formed on the pixel Pb (step S308). Next, the gradation error generated for the pixel Pb is calculated (step S310). Similar to the case of the pixel Pa, the gradation error can be calculated by subtracting the result value from the correction data Cx of the pixel Pb.
[0079]
After the gradation error that has occurred in the pixel of interest Pb is calculated, it is determined whether or not processing has been completed for all the pixels in the processing unit (step S312). A process of diffusing the gradation error that has occurred to these unprocessed pixels is performed (step S314). As described above, since the determination of the pixel Pa in the unit has already been completed, two pixels, the pixel Pc and the pixel Pd, remain as undecided pixels. Therefore, the gradation error generated in the pixel Pb is diffused to these two pixels. The process for diffusing the error from the pixel Pb will be described again with reference to FIG.
[0080]
FIG. 10B is an explanatory diagram conceptually showing a state in which the gradation error generated in the pixel Pb is diffused. In the figure, the pixel Pa is shaded to schematically indicate that the pixel Pa has already been determined for dot formation. At the time when it is determined whether or not dots are formed for the pixel Pb, two pixels, the pixel Pc and the pixel Pd, remain as undetermined pixels in the processing unit. Therefore, the gradation error generated in the pixel Pb is diffused by ½ each to the pixel Pc and the pixel Pd as shown in FIG. In the drawing, white arrows extending from the pixel Pb toward the pixels Pc and Pd schematically indicate that the gradation error generated in the pixel Pb is diffused to these pixels. As a result of the error being diffused in this way, the diffusion error of each pixel is updated to a new diffusion error obtained by adding the diffused error. Here, the gradation error is assumed to be evenly diffused to the undetermined pixels in the processing unit, but of course, the error may be diffused at a predetermined rate instead of being uniformly diffused. .
[0081]
After the gradation error generated in the pixel Pb is diffused to the surrounding pixels, the process returns to step S300 in FIG. 9 again, this time the target pixel is moved to the pixel Pc, and the same process is performed for the target pixel Pc. Hereinafter, such processing will be described with reference to FIG. When the processing for the pixel Pc is started, first, the correction data Cx is calculated by adding the image data DTc of the pixel Pc and the diffusion error EDc. Here, when the dot formation on / off of the pixel Pc is started, the diffusion error EDc of the pixel Pc is added with the error from the pixel Pa and the error from the pixel Pb in addition to the originally stored diffusion error. And updated as a new diffusion error EDc. Therefore, when calculating the correction data Cx of the pixel Pc, a new diffusion error EDc is calculated for the image data DTc of the pixel Pc. By comparing the correction data Cx thus obtained with a predetermined threshold th, it is determined whether or not dots are formed for the pixel Pc. Thus, at the time when it is determined whether or not dots are formed for the pixel Pc, the undetermined pixel remaining in the processing unit is only the pixel Pd as shown in FIG. Therefore, all the gradation errors generated in the pixel Pc are diffused to the pixel Pd, added to the diffusion error originally stored in the pixel Pd, and updated as a new diffusion error EDd.
[0082]
When the gradation error generated in the pixel of interest Pc is diffused, the process returns to step S300 in FIG. 9 to move the pixel of interest to the pixel Pd, and then the determination of dot formation on the pixel Pd is started. That is, the correction data Cx is calculated by adding the new diffusion error EDd associated with the pixel Pd and the image data DTd of the pixel Pd (steps S300 and S302). Next, the correction data Cx is compared with the threshold value th (step S304). If the correction data Cx is larger, it is determined that a dot is formed in the pixel Pd (step S306). Otherwise, the pixel Pd It is determined that no dot is formed (step S308).
[0083]
When it is determined whether or not dots are formed for the pixel Pd in this way, the gradation error generated in the pixel Pd is calculated (step S310), and then it is determined whether or not the processing of all the pixels in the processing unit has been completed (step S312). ). As shown in FIG. 10D, at the time when the processing for the pixel Pd is started, there is no longer any undetermined pixel in the processing unit, and in step S312 after the determination for the pixel Pd is completed. Is determined to have been processed for all the pixels in the processing unit. Therefore, the multi-value process in the unit shown in FIG. 9 is terminated, and the process returns to the gradation number conversion process performed in units as described above with reference to FIG. As shown in FIG. 5, in the unit-unit gradation number conversion process, after returning from the multi-value process in the unit, a process of diffusing the gradation error generated in the processing unit to surrounding units is performed (FIG. 5). 5 step S206).
[0084]
Here, as the value of the gradation error generated in the processing unit, the value of the gradation error obtained for the pixel Pd in step S310 of the intra-unit multilevel processing can be used. This is because the tone error generated in the unit is determined lastly because the tone error generated in the unit is determined while diffusing the tone error generated in the pixel of interest to the undetermined pixels in the unit. Is accumulated as a diffusion error EDd. Hereinafter, this will be described in detail with reference to FIG.
[0085]
The gradation error generated in the pixel Pa is diffused to three pixels of the pixel Pb, the pixel Pc, and the pixel Pd in the processing unit. In determining whether or not the pixel Pb has dot formation, the presence or absence of dot formation is determined so that the error diffused in the pixel Pb is eliminated as much as possible in consideration of the error from the pixel Pa. It is assumed that the error diffused to the pixel Pb can be eliminated by determining the presence or absence of dot formation for the pixel Pb. In this case, a new gradation error does not occur in the pixel Pb, and an error diffused from the pixel Pb to the pixels Pc and Pd does not occur. Similarly, with respect to the other two pixels Pc and Pd, it is assumed that the error diffused in each pixel can be eliminated by determining the presence or absence of dot formation. There is no error. Eventually, in this case, the gradation error generated in the pixel Pa is eliminated in the surrounding undetermined pixels Pb, Pc, and Pd, so that the entire processing unit can avoid the generation of the gradation error. it can.
[0086]
Here, it is assumed that the error in the pixel Pb is eliminated by determining whether or not the dot is formed in the pixel Pb. However, when a new gradation error occurs in the pixel Pb, this error is not detected in the surrounding undetermined pixel Pc. As a result, the gradation error generated in the pixel Pa and the gradation error generated in the pixel Pb are diffused in these pixels. However, if it is possible to eliminate the error diffused to each pixel by determining the presence or absence of dot formation for these pixels Pc and Pd, there will be no gradation error when viewed as a whole unit.
[0087]
Even if it is not possible to eliminate the diffusion error diffused in the pixel Pc by determining the presence or absence of dot formation for the pixel Pc, if the error can be eliminated in the pixel Pd in the same manner, There is no gradation error. In other words, the remaining error that cannot be resolved in the pixel Pd is the gradation error that occurs in the entire unit. As described above, when the presence or absence of dot formation is determined while diffusing the gradation error generated in the target pixel to the undetermined pixels in the unit, the remaining gradation error that cannot be resolved in each pixel is transferred to the subsequent pixels. Since they are collected, the gradation error that has occurred in the pixel Pd that is determined last in the unit can be used as the gradation error that has occurred in the entire processing unit.
[0088]
In the above description, the value of the gradation error generated in the pixel Pd that finally determines the presence or absence of dot formation in the processing unit is used as the gradation error generated in the entire processing unit. The gradation error may be calculated for each pixel constituting the unit, and the sum of these gradation errors may be used as the gradation error of the processing unit.
[0089]
Further, in the example shown in FIG. 10, when determining the presence or absence of dot formation of each pixel constituting the unit of interest, the determination is performed in the order of the pixel Pa, the pixel Pb, the pixel Pc, and the pixel Pd. For example, it may be determined whether or not dots are formed in the order of the pixel Pa, the pixel Pc, the pixel Pb, and the pixel Pd. The order in which the pixel of interest is set in the processing unit can be selected in order of obtaining good image quality.
[0090]
D-3. In-unit multilevel processing of the first embodiment:
As described above, when the gradation number conversion process is performed in units, the same process is repeated for each unit. From this, a pixel that is determined to form a dot in a certain unit tends to be easily determined that a dot is formed at the same position in the subsequent unit. Alternatively, although there is a unit where dots are not formed as individual units, dots may be formed in a fixed pattern with a plurality of units as a period. For example, when a certain area of image data continues, almost the same processing is repeated with a plurality of units as a cycle, and it is possible that dots are periodically formed in pixels at the same position in the unit. Of course, the size of the dot itself is so small that it is difficult to visually recognize. However, if dots are formed in a periodic pattern across multiple units, the existence of such a region may be visually recognized and the image quality may deteriorate. There is.
[0091]
In order to avoid the occurrence of such a problem, in the first embodiment, multi-value processing in the unit is performed, and when image data is converted into dot data in units, dots are formed in the already processed units. The detected pixel position is detected, and the presence or absence of dot formation is determined while reflecting the detection result. In this way, if the presence / absence of dot formation in the processing unit is determined while considering the history of pixel positions where dots are formed in the processed unit, the image quality that can be generated by converting the image data in units of units is determined. Deterioration can be avoided. In the following, the in-unit multilevel processing of the first embodiment will be described.
[0092]
FIG. 11 is a flowchart showing the flow of the intra-unit multilevel processing of the first embodiment. In this process, in contrast to the multi-value processing in the unit described above with reference to FIG. 9, the pixel position where the dot is formed in the processed unit is detected, and the tone number conversion process is performed while reflecting the detection result. The point of going is very different. Hereinafter, the intra-unit multi-value processing of the first embodiment will be described focusing on the differences from the above-mentioned multi-value processing within the unit.
[0093]
Also in the in-unit multi-value processing of the first embodiment, when processing is started, first, one pixel of interest is set from the processing unit (step S400), and correction data Cx for the set pixel of interest is calculated. (Step S402). The correction data Cx can be obtained by adding the image data of the target pixel and the diffusion error diffused and stored in the target pixel. Further, the image data and the diffusion error of each pixel constituting the processing unit have already been read in step S202 during the gradation number conversion process shown in FIG. 5, and data on the target pixel is used from these data. Correction data Cx is calculated.
[0094]
Next, in the intra-unit multi-value process of the first embodiment, a process of detecting a history of pixel positions where dots are formed in a processed unit and setting the threshold th to an appropriate value based on the detection result Is performed (step S404). This process (history detection / threshold setting process) will be described with reference to FIG.
[0095]
FIG. 12 is a flowchart showing the flow of the history detection / threshold setting process. When the history detection / threshold setting processing is started, first, history data reading processing is performed (step S500). The history data is data in which pixel positions determined to form dots for a processed unit are recorded, and stored in a dedicated buffer (history detection buffer) secured in a specific area on the RAM 106. .
[0096]
FIG. 13A is an explanatory diagram conceptually showing the structure of history data. The history data shown in FIG. 13A has a length of 20 bits, and each bit corresponds to a pixel position. As described above, since one unit is composed of four pixels here, five units of pixels are stored in the 20-bit history data. In this example, it is assumed that the data for 5 units is stored in the history data. However, the number of units for storing data is not limited to 5, and data for a larger number of units can be stored. Alternatively, data for a smaller number of units may be stored. Hereinafter, this will be specifically described with reference to FIG.
[0097]
As shown in the figure, the history data can be interpreted as five blocks with four consecutive bits as one block. These five blocks are, in order from the lower side (right side in the figure), the unit processed immediately before the unit being processed ("the previous unit") and the unit processed immediately before ("the previous two units" Each unit corresponds to five units, such as “unit”) and the unit processed immediately before (“three units before”). Also, the four bits constituting each block are, from the lower side, a pixel in the lower right corner (pixel Pd), a pixel in the lower left corner (pixel Pc), and a pixel in the upper right corner (pixel Pb) in the corresponding unit. , Corresponding to the pixel (pixel Pa) in the upper left corner. The fact that a value “1” is recorded in a certain bit in the history data means that a dot is formed in the pixel corresponding to that bit, and conversely the value “0” is stored in the bit. If it is, it means that no dot is formed in the pixel corresponding to the bit.
[0098]
In FIG. 13, the interpretation of 4 bits as one block corresponds to the fact that one unit is composed of four pixels here (see FIG. 6A). Accordingly, if one unit is composed of N pixels, the history data is interpreted as a block of consecutive N bits. Here, the history data is described as 20-bit data. However, one unit is composed of N pixels, and the dot formation status from the processing unit to the unit processed before M units is referred to. In this case, the history data may be data having a length of N × Mbit.
[0099]
As an example, history data as shown in FIG. 13B is assumed. Focusing on a group consisting of four bits on the lower side (right side in the figure) of the history data, the value “1” is set for the most significant (leftmost) bit and the least significant (rightmost) bit in this group. The value “0” is set for the other bits. This is because in the unit processed immediately before the unit being processed (the “previous unit”), dots are formed in the two pixels Pa and Pd, and no dots are formed in the other pixel positions. Means. Similarly, since the value “0” is stored in any bit for the group on the left side in the figure, in the unit processed two times before the processing unit, there is a dot in any pixel. It means that it is not formed. Since the value “1” is stored in the most significant bit in the group further to the left in the figure, the unit processed three times before the processing unit has a dot at the pixel position of the pixel Pa. Is formed. By interpreting the history data in this way, it is possible to know the dot formation status shown in FIG. 13C from the history data assumed in FIG.
[0100]
In step S500 of FIG. 12, such history data is read from the history detection buffer secured in a predetermined area on the RAM 106. After the history data is read out in this way, the process for setting history detection data is started (step S502). Such processing will be described with reference to FIGS.
[0101]
In setting history detection data, first, the number of units for detecting history is set. This is due to the following reason. As described above, in the intra-unit multi-value processing, since it is determined on a unit basis whether or not dots are formed in each pixel, dots may be easily formed at the same pixel position. For example, it is assumed that a region being processed in an image is a relatively low gradation image data region (bright image region) in which dots are formed at a rate of one pixel per 36 pixels. In such a region, dots are formed at a rate of approximately one in a square region composed of a total of 36 pixels of 6 pixels vertically and horizontally. In such an image data area, when it is determined whether or not dots are formed in units of 4 pixels (vertical and horizontal 2 pixels) as one unit, the same processing is repeated with a cycle of approximately 3 units (6 pixels). In other words, dots may be formed at the same pixel position in the third unit.
[0102]
In an area where dots are formed more sparsely, for example, an area where dots are formed at a rate of 1 pixel per 100 pixels, a square area composed of 10 pixels vertically and horizontally has a ratio of approximately one by one. Dots are formed. Therefore, when the presence or absence of dot formation is determined in units of four pixels as one unit, the same processing is repeated with a cycle of 5 units (for 10 pixels), resulting in the formation of dots at the same pixel position in the 5th unit. It can happen.
[0103]
Furthermore, substantially the same can occur even in a region where dots are formed in almost all pixels and pixels where dots are not formed are sparsely generated. For example, in a region where dots are formed at a high density, pixels in which dots are not formed may occur periodically. However, due to the influence of dot bleeding, periodic patterns tend to be less noticeable in regions where dots are formed at a higher density than in regions where dots are formed sparsely. There is a tendency that image quality is not adversely affected without considering the history of as many units as the area.
[0104]
In this way, the period in which dots are formed in a constant pattern depends on the density at which dots are formed, in other words, on the tone value of image data used to determine the presence or absence of dot formation. May be different. In areas where dots are formed sparsely, it is necessary to detect the history of distant units in order to detect patterns with a long period. There is no need to detect history. Therefore, when setting the history detection data, first, how far the history is detected, that is, the number of units to be detected is set.
[0105]
FIG. 14 is an explanatory diagram conceptually showing a state in which the number of units for detecting the history is set according to the gradation value of the image data. As shown in the figure, the number of units for detecting the history is stored in the printer driver in advance in association with the gradation value of the image data. FIG. 14 shows the case where the number of units is stored in association with the sum of the gradation values of each pixel constituting the unit to be processed, but is not limited to this. May be associated with the average value of the gradation values, or stored in association with the gradation value of each pixel constituting the unit.
[0106]
Further, the gradation value Crv shown in FIG. 14 indicates a threshold value that serves as a reference for determining whether or not it is easy to form a dot based on the detected history. That is, as described above, in the area where dots are sparsely formed, dots may be formed periodically, which may adversely affect image quality. Therefore, in order to avoid the occurrence of such a periodic pattern, It is preferable that the dots are not formed periodically. On the other hand, in a region where dots are formed at a high density, there is a possibility that the image quality may be adversely affected by the periodic generation of pixels where dots are not formed. It is preferable that dots are formed in pixels where no dot is formed. That is, when a periodic pattern is generated, the formation of dots is suppressed in the low gradation region and the formation of dots is promoted in the high gradation region. The gradation value Crv shown in FIG. 14 indicates a threshold value for determining whether to suppress or promote dot formation as described above, and when the sum of gradation values is smaller than the threshold value Crv. Dot formation is suppressed, and when the total value is larger than the threshold value Crv, it is determined to promote dot formation.
[0107]
When the number of units for detecting the history is set based on the correspondence shown in FIG. 14, the history detection data is set according to the position of the pixel (target pixel) for which the presence / absence of dot formation is to be determined. As described above, the dot formation may be suppressed or promoted depending on the value of the sum of the gradation values in the unit, but in the following, the sum is less than the threshold value Crv. The case where it is small (when the formation of dots is suppressed) will be described. As an example, consider a case in which it is determined whether or not dots are formed for the pixel Pa in the processing unit. In FIG. 15A, a large square with a broken line indicated by hatching indicates a processing unit, and a black square in the processing unit indicates a pixel of interest.
[0108]
Here, since the gradation value of the image data is small and it is necessary to detect the history from the processing unit to the previous unit, it is necessary to detect whether or not dots are formed in the pixels Pa of these five units. An explanation will be given assuming this. In FIG. 15A, “?” Shown in the pixel Pa of each unit schematically indicates that the dot formation status is detected at these pixel positions.
[0109]
FIG. 15B is an explanatory diagram conceptually showing history detection data set to detect the dot formation status at these pixel positions. The history detection data also has a length of 20 bits, similar to the history data described above, and is interpreted as five groups, with 4 bits as one group. The history detection data is composed of five groups, as described with reference to FIG. 13, corresponding to the fact that the history data stores data for five units. The reason why one group is formed by 4 bits corresponds to the fact that four pixels are collectively processed as one unit in this embodiment. The correspondence between each group of history detection data and the unit, and the correspondence between each bit and the pixel position in the unit are the same as in the case of the history data shown in FIG. . Here, the history detection data is described as data having the same length as the history data shown in FIG. 13, but the present invention is not limited to this.
[0110]
As shown in FIG. 15 (a), since it is assumed here that the dot formation status of the pixel Pa is to be detected for all five units, the history detection data is the pixel of each unit. It is only necessary to set data “1” only for the bit corresponding to Pa and data “0” for the other bits.
[0111]
Further, if the gradation value of the image data is not so small, and the history for three units may be detected, the history detection data may be the data shown in FIG. . That is, for a unit three units before the processing unit, data having a value “1” set in the bit corresponding to the pixel Pa and data “0” set in the other bits may be used.
[0112]
Similarly, when the pixel of interest is set to the pixel Pb, it is only necessary to detect whether or not dots are formed on the pixel Pb of each unit within the range from the processing unit to the set number of units. FIG. 16A is an explanatory diagram showing pixel positions for detecting the dot formation status in each unit when the target pixel is set to the pixel Pb. A pixel displaying “?” In the figure is a pixel that needs to detect the dot formation status. FIG. 16B is an explanatory diagram conceptually showing history detection data set in order to detect the dot formation state in such a pixel. As shown in the figure, history detection data in such a case is data in which the value of the bit corresponding to the pixel Pb of each unit is set to “1” and the other bits are set to the value “0”. Good. If it is sufficient to detect the history for three units, the history detection data may be data as shown in FIG.
[0113]
In step S502 of FIG. 12, as described above, the history detection data is set according to the position of the pixel of interest.
[0114]
When the history detection data is set in this way, the dot detection status (hereinafter referred to as history detection data is set and detected) by applying the set history detection data to the history data read in step S500. The formation status of the dot to be attempted is sometimes referred to as “target history”) (step S504).
[0115]
FIG. 17A is an explanatory diagram conceptually showing how the history detection data is applied to the history data to detect the target history. The upper part of FIG. 17A shows history data, and the lower part of the history data shows history detection data. When detecting the target history, first, a logical product of corresponding bits is calculated between the history data and the history detection data. Next, the target history can be detected by performing a logical sum of the bits of the intermediate data thus obtained. In this specification, “acting history detection data on history data” means that intermediate data is obtained by taking the logical product of corresponding bits between history data and history detection data. Say to calculate. In FIG. 17A, the intermediate data calculated by applying the history detection data to the history data is shown in the lower part of the history detection data. As described above, in the history detection data, the value “1” is set only at the pixel position to be detected in each unit, and the value “0” is set at the pixel position that does not need to be detected. When the logical product of the bits is calculated with the history data and the history detection data, the bit value corresponding to the pixel position is set to “1” only when a dot is formed at the pixel position to be detected. Is done.
[0116]
The above-described processing will be specifically described by paying attention to “the previous unit” shown in FIG. As shown in the history data, dots are formed in the pixel Pa and the pixel Pd in the “previous unit” of the processing unit. Here, since the pixel position where the dot formation status is to be detected is only the pixel Pa as shown in the history detection data, the intermediate data is the pixel Pa as shown in FIG. Only the bit corresponding to “1” has the value “1”, and the other bits have the value “0”. Thus, if any bit of the intermediate data has the value “1”, it means that a dot is formed in the unit at the pixel position to be detected. Thus, by analyzing the value of each bit of the intermediate data, it is possible to detect for each unit whether or not a dot is formed at the pixel position to be detected.
[0117]
In step S504 in FIG. 12, processing for calculating the detection result by taking the logical sum of each bit of the intermediate data is performed as described above. Based on the detection result thus obtained, it is determined whether a target history has been detected (step S506). When the value of the detection result is “0”, it is determined that the target history is not detected. That is, the value of the detection result (that is, the logical sum of each bit of the intermediate data) is “0” because a dot is formed at the pixel position to be detected in any unit to be detected. It means not. Therefore, if the value of the detection result is “0”, it can be determined that the target history has not been detected. On the other hand, if the value of the detection result is “1”, this means that a dot is formed at the pixel position to be detected in any of the units to be detected. Can be determined to be detected.
[0118]
In the intermediate data shown in FIG. 17 (a), the bits in the “1st previous unit” and “3rd previous unit” are set to the value “1”. The value “1” is obtained by calculating the detection result by taking the logical sum of each bit. Therefore, in this case, the target history is detected.
[0119]
As a reference, FIG. 17B shows the dot formation status represented by the history data shown in the upper part of FIG. As shown in FIG. 17A, the detection result calculated from the intermediate data is the value “1”, and a result indicating that a dot is formed in the pixel Pa is obtained. As shown in FIG. 17B, a dot is surely formed at the position of the pixel Pa of “the previous unit” and “the previous unit” of the processing unit. From this, it can be confirmed that the dot formation status can be accurately detected by using the method of applying the history detection data to the history data.
[0120]
In the above description, the case where dots are formed sparsely, that is, the case where the sum of the gradation values in the unit is smaller than the threshold value Crv in FIG. 14 has been described. The case where the total value is larger than the threshold value Crv can be considered in substantially the same manner.
In other words, it was avoided that dots were formed sparsely in a constant pattern in the low gradation area, whereas pixels in which dots were not formed were sparsely formed in a constant pattern in the high gradation area. Although the points to avoid are very different, the others are almost the same. Therefore, in the high gradation region, the history data is once inverted, that is, “0” is set for the bit set to “1”, and “1” is set for the bit set to “0”. As shown in FIG. 17, history detection data may be applied to the data obtained by setting. Thus, if the logical sum of the intermediate data is “1”, it is detected that pixels in which dots are not formed periodically occur.
[0121]
In the history detection / threshold setting processing of FIG. 12, processing for setting the value of the threshold th to an appropriate value is performed based on the target history thus detected. That is, when the target history is not detected in step S506, the normal value Vn is set as the threshold th (step S508).
[0122]
Conversely, when the target history is detected, it is determined whether or not the sum of the gradation values in the unit is larger than the threshold value Crv shown in FIG. 14 (step S510), and the total value is larger than the threshold value Crv. If it is smaller (step S510: no), the premium value Vm set to a value larger than the normal value Vn is set as the threshold th (step S512). Conversely, when the total value is larger than the threshold value Crv (step S510: yes), the additional value Vs set to a value larger than the normal value Vn is set as the threshold value th (step S514). The threshold value th set here is referred to in order to determine whether or not to form a dot on the target pixel during the multi-value processing in the unit. When the threshold value th is set in this way, the history detection / threshold value setting process of FIG. 12 is terminated, and the process returns to the intra-unit multi-value process of the first embodiment shown in FIG.
[0123]
When returning from the history detection / threshold setting process, the magnitude relationship between the correction data Cx calculated in step S402 and the threshold th is determined (step S406). The threshold th is set to an appropriate value by detecting the target history during the history detection / threshold setting process. If the correction data Cx is larger (step S406: yes), it is determined that a dot is formed on the target pixel (step S408). Otherwise (step S406: no), It is determined that no dot is to be formed (step S410). As described with reference to FIG. 12, in the region where the gradation value of the image data is low, when the target history is detected, the value of the threshold th is a predetermined premium value larger than the normally used value (normal value Vn). Since it is changed to (Vm), when the target history is detected, it is difficult to form dots on the target pixel. In the region where the gradation value of the image data is high, when the target history is detected, the value of the threshold th is changed to a predetermined decrease value (Vs) that is smaller than the value that is normally used (normal value Vn). As a result, when the target history is detected, dots are easily formed on the target pixel. The determination result of dot formation presence / absence thus obtained is stored in a variable indicating the determination result for each pixel.
[0124]
When the presence / absence of dot formation for the pixel of interest is determined as described above, the tone error that accompanies the determination is calculated (step S412). The method for calculating the gradation error is the same as that in the unit multi-value processing shown in FIG. Next, it is determined whether or not the processing for all the pixels in the processing unit has been completed (step S414). If unprocessed pixels remain (step S414: no), after performing a process of diffusing an error to those neighboring pixels in the unit (step S416), the process returns to step S400 to move the pixel of interest. Then, the following series of processes is repeated. In step S416, the process of diffusing the error to the pixels in the processing unit can be performed in the same manner as the above-described intra-unit multilevel processing described with reference to FIGS.
[0125]
On the other hand, when it is determined in step S414 that the processing of all the pixels in the processing unit has been completed (step S414: no), a process of writing the determination result for each pixel in the processing unit to the history detection buffer is performed. This is performed (step S418). Such processing will be described with reference to FIGS.
[0126]
FIG. 18 is a flowchart showing the flow of processing for writing the determination result for each pixel of the processing unit into the history detection buffer. When the process of writing to the history detection buffer is started, first, the process of shifting the data stored in the buffer by one unit in the upper direction is performed (step S600). FIG. 19 is an explanatory diagram schematically showing such processing. The small square column shown in the figure schematically shows the history detection buffer. The history detection buffer shown in the upper side of the figure shows the state before the data is shifted. The history detection buffer shown on the lower side shows a state after the data is shifted. In step S600 of FIG. 18, the data in the history detection buffer is shifted upward by one unit as shown in FIG. At this time, the data of the highest unit in the history detection buffer is discarded. In addition, a value “0” is written in each bit of the lowest unit after the shift, just in case. In FIG. 19, the history detection buffer is capable of storing data for five units, but may be capable of storing data for a larger number of units.
[0127]
After the data for one unit is shifted in the upper direction in this way (step S600 in FIG. 18), a process for setting 4-bit length data according to the determination result of each pixel in the processing unit is performed (step S602). Each bit of the data is associated with each pixel constituting the processing unit. Specifically, the most significant bit is associated with the pixel Pa of the processing unit, the second most significant bit is associated with the pixel Pb, the third bit is associated with the pixel Pc, and the least significant bit is associated with the pixel Pd. In step S602, a value “1” is written in a bit corresponding to a pixel in which a dot is formed in the processing unit, and a value “0” is written in a bit corresponding to a pixel in which no dot is formed. For example, when a dot is formed only in the pixel Pb in a certain processing unit and no dot is formed in other pixels, in the processing in step S602, “0100 (2)” is obtained as 4-bit length data corresponding to the processing unit. "Will be set. Note that (2) indicates that a binary number is displayed.
[0128]
In a succeeding step S604, a process for writing the set data in the area of the lowest unit of the history detection buffer is performed. In FIG. 19, the white arrow toward each bit of the lowest unit schematically shows how the set data is written into these bits.
[0129]
As described above, when the process of writing the determination result of each pixel in the processing unit to the history detection buffer is finished, the intra-unit multivalue processing of the first embodiment shown in FIG. Return to the tone number conversion process.
[0130]
As described above, in the intra-unit multilevel processing of the first embodiment, the presence / absence of the target history is detected during the history detection / threshold setting processing, and when the target history is detected, the value of the threshold th is increased. This makes it difficult for dots to be formed on the target pixel. For this reason, even if gradation number conversion processing is performed in units of units, it is possible to avoid a region where dots are formed at the same pixel position with a plurality of units as a cycle, and this region is visually recognized to deteriorate image quality. be able to. Even if dots are not formed on the pixel of interest because the set value of the threshold th is increased, the tone error caused by this is diffused to the surrounding undetermined pixels, and dots are formed at these surrounding pixels. It becomes easy. For this reason, even if it becomes difficult to form dots by detecting the dot formation status in a unit that has already been processed in the pixel where the dots are supposed to be formed, the image quality is reduced due to this. There is no deterioration.
[0131]
In the above description, when the target history is detected, the setting value of the threshold th used for determining whether or not dots are formed is changed to a large value so that it is difficult to form dots on the pixel of interest. Of course, when the target history is detected, it is possible to determine that no dot is formed in the target pixel. In this way, it is possible to easily avoid the formation of dots at the same pixel position with a plurality of units as a cycle.
[0132]
In the above description, history detection data is applied to history data to generate intermediate data, and a logical sum of each bit of the intermediate data is taken to calculate a detection result. On the other hand, the logical product of each bit of the intermediate data is taken, and this value may be used as the detection result. In this case, the value of the detection result is “1” only when dots are formed at all the pixel positions where the dot formation status is to be detected. Therefore, for example, in a situation where dots are formed at a certain density, it is possible to reliably detect only dots with a specific pattern that is easily visible.
[0133]
D-4. Variation:
Below, the modification of the log | history detection and threshold value setting process of 1st Example mentioned above is demonstrated.
[0134]
In the history detection / threshold value setting process of the first embodiment described above, it has been described that the stored history data for 5 units is always read. However, the history data to be read may be made variable according to the number of units to be traced. it can.
[0135]
FIG. 20 is a flowchart illustrating a flow of history detection / threshold setting processing according to the modification. Such processing differs greatly from the processing of the first embodiment described above in that the data length of the history data to be read can be changed according to the number of units for detecting the history. Hereinafter, it demonstrates according to a flowchart centering on difference.
[0136]
When the history detection / threshold setting process of the modified example is started, first, a total value of image data for each pixel constituting the processing unit is calculated (step S700). The image data of each pixel constituting the processing unit is read in advance in step S202 during the gradation number conversion process shown in FIG. 5, and the total value of the image data of each pixel can be easily calculated.
[0137]
Next, the number of units for detecting the history is set in accordance with the correspondence relationship with the total value shown in FIG. 14, and history data having a data length corresponding to the set number of units is read from the above-described history detection buffer (step S702).
[0138]
When the history data is read from the history detection buffer in this way, history detection data is subsequently set (step S704). This process is almost the same as the history detection / threshold value setting process of the first embodiment described above with reference to FIG. 12, except that the bit length of the history detection data varies depending on the processing unit. That is, in the above-described embodiment, the value “0” is set to all the pixels for the unit that does not detect the history in the history data. However, in this modification, the history data is stored by the unit for detecting the history. Since the data is read out, the data length of the history detection data varies according to the processing unit.
[0139]
Next, history detection data is applied to the history data to detect the presence or absence of the target history (the formation status of the dot to be detected). When the target history is detected in a region where the total sum of the tone values in the unit is smaller than the threshold value Crv shown in FIG. 14, the threshold value th is changed to a large value (Vm), and the total value is the threshold value. When the target history is detected in an area larger than Crv, the value of the threshold th is changed to a small value (Vs). If the target history is not detected, the threshold value th is set to the normal value Vn (steps S706 to S716). When the above processing is completed, the history detection / threshold value setting processing of the modified example is exited, and the processing returns to the in-unit multi-value processing shown in FIG.
[0140]
According to the history detection / threshold value setting process of the modified example described above, only necessary data can be read out from the stored history data, so that the process can be speeded up.
[0141]
E. Second embodiment:
In the history detection / threshold setting processing performed during the unit multi-value processing of the first embodiment described above, history data representing the dot formation status of the processed unit is stored in one history detection buffer. It was explained as being. Since the unit includes a plurality of pixels, in the first embodiment, data indicating the presence or absence of dot formation for each pixel is mixedly stored on the history detection buffer. On the other hand, as many buffers as the number of pixels constituting the unit are provided, and it is determined whether or not dots are formed for each pixel, the determination results may be stored in different buffers in association with the pixel positions in the unit. . Hereinafter, a second embodiment using a dedicated buffer for each pixel position will be described.
[0142]
E-1. In-unit multi-value processing of the second embodiment:
FIG. 21 is a flowchart showing the flow of the intra-unit multi-value processing of the second embodiment. This processing is largely different from the intra-unit multilevel processing of the first embodiment described above with reference to FIG. 11 in that a history detection buffer is provided for each pixel position. Another difference from the first embodiment is that the determination result is written in the corresponding history detection buffer every time it is determined whether or not a dot is formed for the target pixel. . In the following, the in-unit multi-value processing of the second embodiment will be described focusing on these differences from the first embodiment.
[0143]
When the process is started, first, a target pixel from which a dot is formed is determined from a plurality of pixels constituting the processing unit (step S800), and image data and diffusion error of the target pixel are detected. The correction data Cx is calculated by addition (step S802). As the image data and diffusion error for each pixel constituting the processing unit, data already read during the gradation number conversion processing shown in FIG. 5 is used.
[0144]
When the correction data Cx for the pixel of interest is thus calculated, the history detection / threshold setting process of the second embodiment is started (step S804). As described above, in the second embodiment, a dedicated history detection buffer provided according to the pixel position in the unit is used. Therefore, before describing the contents of such processing, the history detection buffer used in the history detection / threshold setting processing of the second embodiment will be described.
[0145]
FIG. 22 is an explanatory diagram schematically showing a history detection buffer provided in association with the pixel position in the unit. The “history detection buffer A” shown in the figure is a buffer associated with the pixel Pa in the unit. This buffer stores the dot formation determination result for the pixel Pa. The “history detection buffer B” is a buffer in which the determination result for the pixel Pb is stored in association with the pixel Pb in the unit. Similarly, the determination result for the pixel Pc in the unit is recorded in the “history detection buffer C”, and the determination result for the pixel Pd is stored in the “history detection buffer D”. Thus, in the second embodiment, dedicated buffers associated with pixel positions in the unit are provided, and history data is stored in the respective buffers.
[0146]
Next, taking the “history detection buffer A” as an example, the data structure of the stored history data will be described. Each bit of the history detection buffer A is associated with the previous unit one by one from the processing unit in order from the lower side (right side in FIG. 22). That is, the least significant bit (the rightmost bit in the figure) is associated with the “one previous unit” of the processing unit, and the second bit from the lower level is assigned to the “second previous unit” and the lower 3 bits. Each bit is associated with each unit in order, for example, the second bit is “3rd previous unit”. The value “1” being set in these bits indicates that a dot is formed in the pixel Pa of the corresponding unit, and conversely that the value “0” is set corresponds. This indicates that no dot is formed on the pixel Pa of the unit.
[0147]
As is clear from the above description, the history data illustrated in FIG. 22 stored in the history detection buffer A is for the units one, three, five, eight, and eight before the processing unit. Indicates that a dot is formed in the pixel Pa and no dot is formed in the pixel Pa for other units. In FIG. 22, each history detection buffer has a data length of 8 bits, but it may be data having a larger number of bits.
[0148]
The history detection / threshold setting process of the second embodiment performed using such history detection data will be described below. The history detection / threshold setting process of the second embodiment is different from the process of the first embodiment described above in that a history detection buffer is provided for each pixel position, and the other processes are almost the same. It is. Accordingly, the following description will be made with reference to the flowchart of FIG. 12 describing the history detection / threshold value setting process of the first embodiment. This process is a process executed in step S804 on the target pixel after the target pixel is set in the in-unit multi-value processing of FIG. 21 (step S800).
[0149]
When the history detection / threshold setting processing of the second embodiment is started, first, history data is read from the history detection buffer (corresponding to step S500 in FIG. 12). As described with reference to FIG. 22, in the second embodiment, a history detection buffer is provided according to the pixel position, and in this process, a necessary data amount is obtained from the buffer corresponding to the same pixel position as the target pixel. Read the history data. For example, assuming that the pixel of interest is located in the pixel Pa and the dot formation status is detected retroactively by 5 units, data of 5 bits is read from the lower side of the “history detection buffer A”. As a result, in the example shown in FIG. 22, “10101 (2)” is read as the history data. Here, (2) indicates that a binary number is displayed. If the target pixel is set to the pixel Pb, the pixel Pc, and the pixel Pd, the history data recorded in the history detection buffer B, the history detection buffer C, and the history detection buffer D are read out. Good.
[0150]
When the history data is read in this way, subsequently, processing for setting history detection data is performed (corresponding to step S502 in FIG. 12). Such processing will be described with reference to FIG. For example, it is assumed that the dot formation status is detected retroactively from the processing unit to the previous unit. FIG. 23A is an explanatory diagram conceptually showing the distribution of pixel positions for which the dot formation status is to be detected. Each small square in the drawing schematically shows a pixel. A large square indicated by a thick broken line schematically shows a unit being processed, and a large square indicated by a thin broken line schematically shows a unit that has already been processed. The hatched pixels in the processing unit indicate the target pixel, and the pixels marked with “?” In the processed unit indicate the pixels for which the dot formation status is to be detected.
[0151]
FIG. 23B shows history detection data used to detect the dot formation status for such pixels. Here, since it is assumed that detection is performed retroactively from the processing unit to the previous unit, the value “1” is set in the bits from the least significant bit to the 5th bit in the history detection data, and the higher bit than the 6th bit is set. Is set to the value “0”. In this way, history detection data can be set by setting a value “1” in the bit corresponding to the unit to be detected and setting a value “0” in the other bits. FIG. 23C shows the history detection data set in this way displayed in hexadecimal.
[0152]
When history detection data is set in this way, the target history (the dot formation status at the pixel position to be detected) is detected by applying the set history detection data to the previously read history data (FIG. 12). Step S504). The target history is detected in the same manner as in the first embodiment. That is, the intermediate data is calculated by taking the logical product of the corresponding bits between the history data and the history detection data, and the detection result is calculated by taking the logical sum of each bit of the obtained intermediate data. . If the obtained detection result is the value “1”, it is determined that the target history has been detected, and the premium value Vp is set to the threshold th. If the detection result is the value “0”, it is determined that the target history has not been detected, and the normal value Vn is set as the threshold th. When the threshold value th is set to an appropriate value as described above, the history detection / threshold value setting process of the second embodiment is terminated, and the process returns to the in-unit multi-value process shown in FIG. Here, the detection result is calculated by taking the logical sum of each bit of the intermediate data, but the detection result may be calculated by taking the logical product of each bit of the intermediate data.
[0153]
When the threshold value th is set as described above and the processing returns to the multi-value processing in the unit of the second embodiment shown in FIG. 21, the magnitude relationship between the correction data Cx of the target pixel and the threshold value th is subsequently determined. (Step S806 in FIG. 21). If the correction data Cx is larger, it is determined that a dot is to be formed on the pixel of interest (step S808). Otherwise, it is determined that no dot is to be formed (step S810). Next, after calculating a gradation error caused by determining the presence or absence of dot formation (step S812), a process of writing the determination result for the pixel of interest into the corresponding history detection buffer is started (step S814). Such processing will be described with reference to FIG.
[0154]
FIG. 24 is an explanatory diagram conceptually showing a state in which the determination result for the pixel of interest is written in the history detection buffer in the unit multi-value quantization process of the second embodiment. Assuming that the pixel of interest is set to the pixel Pa, the determination result of dot formation presence / absence is written in the history detection buffer A. In writing the determination result, the history data stored in the history detection buffer A is first shifted to the upper side (left side in the figure) by 1 bit. The upper part of FIG. 24 shows the history data stored in the history detection buffer A before shifting, and the lower part shows the history data stored in the buffer after shifting. As the history data is shifted, the data stored in the most significant bit is discarded. Next, a process of writing a value corresponding to the determination result for the pixel of interest in the least significant bit is performed. If it is determined that a dot is to be formed in the pixel of interest, the value “1” is written in the least significant bit, and if it is determined that no dot is to be formed in the pixel of interest, a value “0” is written. In the above description, the pixel of interest has been described as being set to the pixel Pa. However, the history detection corresponding to the pixel position of the pixel of interest is similarly performed when the pixel of interest is set to another pixel position. A value corresponding to the determination result may be written in the buffer. In step S814 in FIG. 21, a process for writing the determination result for the target pixel in this way into the history detection buffer corresponding to the target pixel is performed.
[0155]
When the process of writing the determination result of the target pixel is completed in this way (step S814 in FIG. 21), it is determined whether the process for all the pixels in the processing unit is completed (step S816), and unprocessed pixels remain. If so, the process returns to step S800 to repeat the series of processes described above. When the process for all the pixels is completed, the process returns to the gradation number conversion process shown in FIG.
[0156]
In the intra-unit multilevel processing of the second embodiment described above, each bit of the history detection buffer is associated with each processed unit. Accordingly, even when the number of units to be examined retrospectively for the dot formation status increases, it is only necessary to read data having the same bit length as the number of units. For example, when detecting the dot formation status by going back 8 units from the unit being processed, it is necessary to read the history data of 32 bits in the first embodiment, assuming that one unit is composed of four pixels. However, in the second embodiment, it is only necessary to read 8-bit data. For this reason, in the second embodiment, even if the number of units to be traced back is large, a quick process can be performed.
[0157]
E-2. Modification of the second embodiment:
In the second embodiment described above, the pixel position for detecting the dot formation state is described as being the same pixel as the pixel of interest. However, similarly to the first embodiment, the pixel position for detecting the dot formation status can be freely set. In the second embodiment, the pixel for detecting the dot formation status is also a unit. It is possible to set an arbitrary pixel position every time. Hereinafter, a modification of the second embodiment will be described.
[0158]
FIG. 25 is an explanatory diagram showing a method of detecting the dot formation status at an arbitrary pixel position in the second embodiment. FIG. 25A is an explanatory diagram illustrating the distribution of pixel positions to be detected. The small squares shown in the figure schematically indicate pixels, and the large squares indicated by broken lines indicate units. The pixel position where the dot formation state is to be detected is indicated by a “?” Mark in the pixel. Although the pixel position and the number of pixels to be detected can be set arbitrarily, the following description will be made assuming that the dot formation status at the pixel position shown in FIG.
[0159]
FIG. 25B is an explanatory diagram illustrating a method for detecting a dot formation state at an arbitrary pixel position in the second embodiment. In the second embodiment, as described above, since the determination result of dot formation presence / absence is stored for each pixel position of each unit, each dot formation state at an arbitrary pixel position can be detected. Detection is performed by a method focusing on the pixel position of the unit. In other words, in the example shown in FIG. 25A, the dot formation status for the pixel position of the pixel Pa is to be detected by the “one previous unit”, “two previous units” of the processing units, These are the three units of “3rd previous unit”. Further, the pixel position of the pixel Pb is two units of “the previous unit” and “the second previous unit” of the processing unit. Furthermore, as for the pixel position of the pixel Pc or the pixel Pd, the dot formation status is detected only in the “one unit before” of the processing unit. In this way, after detecting the unit position for detecting the dot formation status for each pixel position, if it is determined for each pixel position whether or not a dot is formed in the detected unit, any pixel position can be determined. It is possible to detect the dot formation status. Hereinafter, specific description will be made in accordance with the case shown in FIG.
[0160]
First, the case where the dot formation status of the pixel Pa is detected will be described. FIG. 25B schematically shows how the dot formation status at the pixel position of the pixel Pa is detected in this way. First, the history data for 3 units (data corresponding to 3 bits from the lowest order) is read from the buffer (history detection buffer A) in which the dot formation determination result at the pixel Pa is stored. Of course, when detecting a larger number of units retrospectively, the data length to be read out may be increased accordingly.
[0161]
Next, history detection data is applied to the history data. The history detection data is set according to the unit that is trying to detect the dot formation status at the pixel Pa. As shown in FIG. 25 (a), for the pixel Pa, the dot formation status is detected by three units of the processing unit "one previous unit", "second previous unit", and "three previous units". In response to this, in the history detection data, the value “1” is set in the three least significant bits. The history detection data set in this way ("111 (2)" in binary notation) is applied to the read history data to calculate intermediate data. “Making history detection data act on history data” means taking a logical product between corresponding bits of two data as described above. When the intermediate data is obtained in this way, a detection result for the pixel Pa can be obtained by calculating a logical sum between each bit of the intermediate data.
[0162]
The detection result thus obtained for the pixel Pa means the following. That is, the detection result having the value “0” means that no dot is formed in the pixel Pa in any of the units to be detected, and conversely, the detection result is “1”. This means that a dot is formed in the pixel Pa in any of the units to be detected. Of course, instead of taking the logical sum of each bit of the intermediate data, a logical product may be taken. When the logical product of each bit of the intermediate data is taken, the detection result value of “1” means that dots are formed in the pixels Pa of all the units to be detected. Become.
[0163]
The detection result for the pixel Pb can be obtained by the same method. FIG. 25C schematically shows how the dot formation status for the pixel Pb is detected. Briefly described with reference to FIG. 25 (b), first, history data for three units is read from a buffer (history detection buffer B) in which the presence or absence of dot formation for the pixel Pb is recorded. Next, history detection data is applied to this data to calculate intermediate data. For the pixel Pb, since the dot formation status of the “one previous unit” and “two previous units” of the processing unit is to be detected, the history detection data corresponding to this is the least significant bit. From which two bits are set to the value “1” (“011 (2)” in binary notation). Thus, by obtaining a logical sum between each bit of the obtained intermediate data, a detection result for the pixel Pb can be obtained.
[0164]
The detection result for each pixel can be calculated in the same manner for the pixels Pc and Pd. For these pixels, the unit that is trying to detect the dot formation status is only “the previous unit” of the processing unit. Therefore, for these pixels, the value “1” is set only for the least significant bit as history detection data. May be used (“001 (2)” in binary display). FIG. 25 (d) shows how the detection result for the pixel Pc is calculated, and FIG. 25 (e) shows how the detection result for the pixel Pd is calculated.
[0165]
Once the detection results for each pixel are obtained in this way, the final detection result can be obtained by calculating the logical sum of these detection results. That is, if a dot is formed in any of the pixels indicated by “?” In FIG. 25A, a value “1” is obtained as a detection result, and conversely, a dot is formed in any pixel. If not, the value “0” is obtained as the detection result. Of course, the final detection result may be calculated by taking the logical product of the detection results for each pixel. In such a case, when dots are formed in all the pixels indicated by “?” In FIG. 25A, the final detection result value is “1”.
[0166]
By using the method described above, it is possible to freely set a pixel position from which a dot formation state is to be detected from among already processed units. Therefore, even in the case where the presence / absence of dot formation is determined on a unit basis, the dot formation status at a predetermined pixel position is detected, and the presence / absence of dot formation is determined based on the detection result. It is possible to effectively avoid the formation of dots in a predetermined pattern.
[0167]
While various embodiments have been described above, the present invention is not limited to all the embodiments described above, and can be implemented in various modes without departing from the scope of the invention.
[0168]
For example, in the various embodiments described above, a so-called error diffusion method is used as a method for determining the presence or absence of dot formation for each pixel constituting the unit. However, the present invention is not limited to this, and other known methods such as a dither method can be used. Techniques can also be applied.
[0169]
A software program (application program) that realizes the above-described functions may be supplied to a main memory or an external storage device of a computer system via a communication line and executed. Of course, a software program stored in a CD-ROM or a flexible disk may be read and executed.
[0170]
In the various embodiments described above, the image data conversion process including the gradation number conversion process has been described as being executed in the computer. However, part or all of the image data conversion process is performed on the printer side or a dedicated image processing apparatus. It may be executed by using.
[0171]
Furthermore, the image display device is not necessarily limited to a printing device that prints an image by forming ink dots on a print medium. For example, bright spots are dispersed at an appropriate density on a liquid crystal display screen. Thus, a liquid crystal display device that expresses an image whose gradation changes continuously may be used.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram of a printing system illustrating an outline of the present invention.
FIG. 2 is an explanatory diagram illustrating a configuration of a computer as an image processing apparatus according to the present exemplary embodiment.
FIG. 3 is a schematic configuration diagram of a printer as an image display apparatus according to the present exemplary embodiment.
FIG. 4 is a flowchart illustrating a flow of image data conversion processing performed by the image processing apparatus according to the present exemplary embodiment.
FIG. 5 is a flowchart showing a flow of gradation number conversion processing performed in units.
FIG. 6 is an explanatory diagram showing a state in which a plurality of pixels are collectively set as a processing unit.
FIG. 7 is an explanatory view exemplifying a state in which a gradation error generated in a processing unit is diffused to surrounding undetermined pixels.
FIG. 8 is an explanatory diagram exemplifying a ratio of diffusing a gradation error generated in a processing unit to surrounding undetermined pixels.
FIG. 9 is a flowchart showing a flow of a process for multi-value the inside of a unit by determining the presence or absence of dot formation for each pixel constituting the unit.
FIG. 10 is an explanatory diagram conceptually showing a method of determining the presence or absence of dot formation for each pixel while diffusing an error to each pixel in the unit.
FIG. 11 is a flowchart showing a flow of processing for multi-valued in the unit by determining the presence or absence of dot formation for each pixel in the unit in the first embodiment.
FIG. 12 is a flowchart showing a flow of processing for detecting a dot formation state and setting a threshold value in the first embodiment.
FIG. 13 is an explanatory diagram conceptually showing the structure of history data read in the history detection / threshold value setting process of the first embodiment.
FIG. 14 is an explanatory diagram conceptually showing a state in which the number of units for detecting a history is set in association with the sum of the gradation values of the pixels constituting the unit.
FIG. 15 is an explanatory diagram conceptually showing the structure of history detection data set for detecting a history in the history detection / threshold value setting process of the first embodiment;
FIG. 16 is an explanatory diagram conceptually showing the structure of different history detection data.
FIG. 17 is an explanatory diagram conceptually showing a state of detecting a target history by applying history detection data to history data in history detection / threshold setting processing of the first embodiment.
FIG. 18 is a flowchart showing a flow of processing for writing a determination result for each pixel in a unit to a history detection buffer storing history data in the first embodiment.
FIG. 19 is an explanatory diagram conceptually showing a state in which a determination result for each pixel in a unit is written in a history detection buffer storing history data in the first embodiment.
FIG. 20 is a flowchart showing a flow of processing for detecting a dot formation state and setting a threshold value in a modification of the first embodiment.
FIG. 21 is a flowchart showing a flow of processing for multi-valued in the unit by determining whether or not dots are formed for each pixel in the unit in the second embodiment.
FIG. 22 is an explanatory diagram conceptually showing a history detection buffer provided in association with pixel positions in a unit in the second embodiment.
FIG. 23 is an explanatory diagram conceptually showing a method of setting history detection data in accordance with the pixel position where the dot formation status is to be detected in the second embodiment.
FIG. 24 is an explanatory diagram conceptually showing a method of writing a determination result of dot formation presence / absence in a history detection buffer in the second embodiment.
FIG. 25 is an explanatory diagram conceptually showing a method for detecting a dot formation state at an arbitrary pixel position by applying history detection data to history data in the second embodiment.
[Explanation of symbols]
10 ... Computer
12. Printer driver
14 ... Tone conversion module
20 Color printer
100: Computer
102 ... CPU
104 ... ROM
106 ... RAM
108 ... Peripheral device interface P / IF
109 ... Disk controller DDC
110: Network interface card NIC
112 ... Video interface V / IF
114 ... CRT
116 ... Bus
118: Hard disk
120 ... Digital camera
122 ... Color scanner
124: Flexible disk
126 ... Compact disc
200 ... Color printer
230 ... Carriage motor
235 ... Paper feed motor
236 ... Platen
240 ... carriage
241 ... Print head
242, 243 ... Ink cartridge
244 ... Ink ejection head
260 ... control circuit
261 ... CPU
262 ... ROM
263 ... RAM
300 ... communication line
310 ... Storage device

Claims (7)

An image processing apparatus that converts image data represented by gradation values for each pixel constituting an image into dot data indicating the presence or absence of dot formation for each pixel,
Dot data conversion means for grouping a predetermined plurality of adjacent pixels as a unit, and converting the image data into the dot data in units.
For at least some of the pixels constituting the converted unit, comprising dot presence / absence storage means for storing the presence / absence of dot formation in a state where the pixel positions in the unit can be identified, and
The dot data conversion means includes
Converted unit selection means for calculating a total sum of gradation values for each pixel constituting the unit to be converted to dot data and selecting a predetermined number of units from the converted units based on the total value When,
A pixel-of-interest setting means for sequentially setting a pixel of interest for determining the presence or absence of dot formation within the dot data conversion target unit;
Dot formation presence / absence detection means for detecting, with respect to the selected predetermined number of units, dot formation presence / absence in a pixel at a predetermined position in the converted unit corresponding to the pixel position of the target pixel;
Based on the gradation value of the image data for the pixel of interest and the presence / absence of the detected dot formation, the dot formation of the pixel of interest is made easier as the gradation value of the pixel of interest increases. Dot formation determination means for determining presence or absence,
When the converted unit selection means is a sum value smaller than a first predetermined value, the converted unit selection means is more than the sum value larger than the first predetermined value and smaller than a second predetermined value. An image processing apparatus as means for selecting a unit far from the pixel of interest. The image processing apparatus according to claim 1,
The converted unit selection means includes:
When the sum is smaller than the first predetermined value, a unit that is closer to the farthest unit when the sum is larger than the first predetermined value and smaller than the second predetermined value, An image processing apparatus, which is a means for storing in association with each other. A printing apparatus that receives image data represented by gradation values for each pixel constituting an image and prints the image by forming dots on a print medium based on the image data,
Dot data conversion means for grouping a predetermined plurality of adjacent pixels as a unit, and converting the image data into the dot data in units.
For at least some of the pixels constituting the converted unit, dot presence / absence storage means for storing the presence / absence of dot formation in a state where the pixel position in the unit can be identified;
Dot forming means for forming dots on each of the pixels based on the obtained dot data,
The dot data conversion means includes
A converted unit that calculates a total sum of gradation values for each pixel constituting a unit to be converted into the dot data, and selects a predetermined number of units from the converted units based on the total value A selection means;
A pixel-of-interest setting means for sequentially setting a pixel of interest for determining the presence or absence of dot formation in the unit to be converted into the dot data;
Dot formation presence / absence detection means for detecting, with respect to the selected predetermined number of units, dot formation presence / absence in a pixel at a predetermined position in the converted unit corresponding to the pixel position of the target pixel;
Based on the gradation value of the image data for the pixel of interest and the presence / absence of the detected dot formation, the dot formation of the pixel of interest is made easier as the gradation value of the pixel of interest increases. Dot formation determination means for determining presence or absence,
When the converted unit selection means is a sum value smaller than a first predetermined value, the converted unit selection means is more than the sum value larger than the first predetermined value and smaller than a second predetermined value. A printing apparatus that is means for selecting a unit far from the pixel of interest. The printing apparatus according to claim 3,
  The converted unit selection means includes:
    If the total value is smaller than the first predetermined value, the unit that is closer to the farthest unit when the total value is larger than the first predetermined value and smaller than the second predetermined value, A printing apparatus, which is a means for storing in association with each other. An image processing method for converting image data expressed by gradation values for each pixel constituting an image into dot data indicating the presence or absence of dot formation for each pixel,
A first step of grouping a predetermined plurality of pixels adjacent to each other as a unit, and converting the image data into the dot data in units;
For at least some of the pixels constituting the converted unit, a second step of storing the presence or absence of dot formation in a state where the pixel positions in the unit can be identified, and
The first step includes
A sub-step of selecting a predetermined number of units from the converted units according to the correspondence with the total value based on the gradation value for each pixel constituting the unit to be converted into the dot data ( 1-a)
A sub-step (1-b) for sequentially setting a pixel of interest for determining whether or not to form dots in a unit to be converted into dot data;
A sub-step (1-c) for detecting, with respect to the selected predetermined number of units, the presence or absence of dot formation at a pixel at a predetermined position in the converted unit corresponding to the pixel position of the target pixel;
Based on the gradation value of the image data for the pixel of interest and the presence / absence of the detected dot formation, the dot formation of the pixel of interest is made easier as the gradation value of the pixel of interest increases. And sub-step (1-d) for determining the presence or absence,
In the lower step (1-a), when the total value is smaller than the first predetermined value as the correspondence relationship with the total value, the lower step (1-a) is larger than the first predetermined value and larger than the second predetermined value. An image processing method, which is a step of selecting a unit farther from the pixel of interest than when the sum is smaller. The image processing method according to claim 5, comprising:
  The substep (1-a) includes
    If the sum is smaller than the first predetermined value, a unit closer to the farthest unit when the sum is larger than the first predetermined value and smaller than the second predetermined value is also selected. An image processing method which is a process to perform. A program for realizing, using a computer, a method for converting image data expressed by gradation values for each pixel constituting an image into dot data indicating the presence or absence of dot formation for each pixel. ,
A first function of grouping a plurality of adjacent pixels as a unit, and converting the image data into the dot data in units;
For at least some of the pixels constituting the converted unit, the second function of storing the presence / absence of dot formation in a state where the pixel positions in the unit can be identified, and
The first function further includes:
A function (1) for selecting a predetermined number of units from the converted units based on the gradation value for each pixel constituting the unit to be converted into the dot data, according to the correspondence relationship with the total value. -A)
A function (1-b) for sequentially setting a pixel of interest for determining the presence or absence of dot formation in the unit to be converted into the dot data;
A function (1-c) for detecting, with respect to the selected predetermined number of units, the presence or absence of dot formation at a pixel at a predetermined position in the converted unit corresponding to the pixel position of the target pixel;
Based on the gradation value of the image data for the pixel of interest and the presence / absence of the detected dot formation, the dot formation of the pixel of interest is made easier as the gradation value of the pixel of interest increases. And a function (1-d) for determining presence or absence,
When the function (1-a) is a total value smaller than the first predetermined value as a correspondence relationship with the total value, the function (1-a) is larger than the first predetermined value and larger than the second predetermined value. A program for selecting a unit farther from the pixel of interest than when the sum is smaller.

Images