JP2006148334A - Image processing apparatus and recording system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus and a recording system capable of obtaining a uniform image without a granular texture in the case that a plurality of kinds of multi-value signals having two kinds or more of elements such as "dot diameter" and "color" are quantized and a prescribed image is expressed by pseudo halftone processing. <P>SOLUTION: The image processing apparatus adds a plurality of kinds of multi-value signals by ink colors (or by dot diameters) to acquire a plurality of addition values and thereafter refers to a multi-dimensional LUT to acquire a tentative quantization value by the ink colors (or by the dot diameters). Thereafter the image processing apparatus separates each of the tentative quantization values by the ink colors (or by the dot diameters) to acquire a final quantization value. Since the image processing apparatus can distribute each of a plurality of kinds of dots in a proper state even in an output image obtained by overlapping a plurality of planes in which a plurality of elements are mixed, the image processing apparatus can output a uniform image without a granular texture. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image processing method and a recording system for realizing the method, and more particularly to a quantization method for expressing gradation of each color by pseudo gradation processing.

  In general, when a color image is recorded by a recording apparatus, a quantization process is performed to convert multi-value data (for example, 8-bit 256 gradations) given to each pixel to a lower level. For example, in the case of a recording apparatus that expresses density with binary values such as dot recording / non-recording, an error diffusion method that converts multi-value data into binary data is generally known as quantization processing ( (Refer nonpatent literature 1.). The error diffusion method will be briefly described below.

  For example, in the pixel coordinates in the image, the density given to a certain pixel of interest P (x, y) is V (x, y), and the threshold for binarization is T. In this case, if V (x, y) is larger than T, the target pixel is determined to be recorded (1: highest density), and if V (x, y) is smaller than T, the target pixel is determined not to be recorded (0: lowest density). However, the actual V (x, y) is almost always an intermediate density that is neither the highest density nor the lowest density. Therefore, an error E (x, y) occurs between the input intermediate density V (x, y) and the output density (the highest density or the lowest density 0). In the error diffusion method, this error E (x, y) is converted into peripheral pixels of the target pixel, that is, P (x + 1, y), P (x-1, y + 1), P (x, y + 1), P (x + 1, y + 1). Sort to etc. The distribution method is not particularly defined, but is generally distributed to the surroundings while being weighted according to empirically obtained coefficients W0, W1, W2, and W3. The distributed value is added to the original multi-value data and compared with the threshold value as a new V (x, y).

  In the error diffusion method, recording or non-recording is determined for all pixels by repeating the above operation. In the resulting image, the density of the original image is substantially realized by the average density around the target pixel. According to such an error diffusion method, recording dots are relatively excellent in dispersibility, and a smooth image can be obtained.

  In addition to the error diffusion method, a binarization method using a mask with blue noise characteristics is also proposed in order to place importance on the visibility of the dot arrangement and to arrange multiple dots in a highly dispersive state as much as possible. (See Patent Document 1). In any case, by adopting a binarization method that can obtain a state of relatively high dispersibility in this way, it has been possible to reduce graininess and output a smooth and high-quality image. .

  By the way, when recording a full-color image with a plurality of colors of ink, it is confirmed that the smoothness of the image is impaired only by simply applying the error diffusion method described in Non-Patent Document 1 or the method described in Patent Document 1. It was done. Specifically, when an image for one plane is formed using one color ink, a smooth image can be obtained by adopting the method described in the above document, but a plurality of color inks can be obtained. When a plurality of planes are overlaid using the, new noises, particularly noises with poor visibility, are generated by the superposition of the planes.

  In order to solve such a problem, proposals have been made in consideration of dot arrangements between a plurality of colors (for example, see Patent Document 2 and Patent Document 3). According to these patent documents, an example focusing on the arrangement of cyan dots and magenta dots that particularly affect the visibility is given, and the state where both are overlapped as much as possible and visually pleasant when they overlap A binarization method is disclosed in which error diffusion processing or blue noise is applied to the sum of the input signal values of the two so as to be dispersed.

  Furthermore, in order to realize the above Patent Document 2 or 3, paying attention to the need for large hardware or software having a large capacity memory, the Patent Document 2 or Patent Document can be constructed with a simpler configuration. 3 has also been proposed (see Patent Document 4). According to Patent Document 4, for each input value of two components (signals) such as cyan and magenta, for example. A configuration is disclosed that realizes a dot arrangement in which dots do not overlap as much as possible by referring to a two-dimensional lookup table (LUT). Furthermore, in Patent Document 4, applicable LUTs are not limited to two dimensions. For example, a three-dimensional LUT that takes into account three components obtained by adding yellow to cyan and magenta, and a LUT with more dimensions. Is clearly stated to be effective.

  As described above, in recent color image recording apparatuses, a device for outputting a smooth and high-quality image while considering the dispersibility of dots for each color and the dispersibility of dots when multiple colors are superimposed. Was given.

Japanese Patent No. 2,622,429 JP-A-8-279920 Japanese Patent Laid-Open No. 11-10918 JP 2003-1116014 A An Adaptive Algorithm for Spatial Gray Scale "in society for Information Display 1975 Symposium Digest of Technical Papers, 1975, 36

  By the way, in recent years, in order to output an image with higher gradation, the dot diameter can be adjusted in several stages, or multiple types of inks having the same color but different densities can be recorded. Many recording devices that can be adjusted in several stages are provided. In the case of such a recording apparatus, not the binary error diffusion method described above but a multi-value error diffusion method to which this is applied can be applied. In the multilevel error diffusion method, as in the binary error diffusion method, the error generated in the pixel of interest is distributed to the surroundings, but the output value is not binary, and there are several stages that can be expressed by the printing apparatus. It is a value.

  The second embodiment of Patent Document 4 also describes a configuration in which an effect peculiar to the document can be obtained using a recording apparatus in which the density of each pixel can be adjusted in a plurality of stages. For example, by referring to a two-dimensional lookup table (LUT) for each input value of two components such as cyan and magenta, the dots of each other are not overlapped as much as possible, and the individual ink colors are A quantization method having multiple signal values is disclosed. Even if the same cyan is used, when a recording head capable of recording by switching between large dots and small dots is prepared, each pixel after quantization is expressed in gradations from 0 to 2, and in the case of 0, dots are represented. In the case of non-recording, 1 is controlled by recording with small dots, and in the case of 2, control is performed as recording by large dots. As described above, in the multi-value recording disclosed in Patent Document 4, the types of dots to be recorded in the pixel are defined from the signal values of several stages finally given to the pixel.

  However, in recent years, the types of recording media to be supported have increased and their characteristics have also varied. Even when trying to express the same gradation value, the number of required large dots (dark ink) and the number of small dots (light ink), or their ratio, may vary depending on the type of recording medium. Will be different. That is, it is necessary to appropriately adjust the number of large dots (dark ink) and the number of small dots (light ink) according to the type of recording medium. However, as in Patent Document 4, for example, in the final stage after quantization to 2 bits (3-value or 4-value), the ink is divided into two types of ink (dark and light, or large and small). However, there were strong restrictions on the way of sorting, and fine density adjustment could not be performed.

  Therefore, in a recent ink jet recording apparatus, large dots and small dots (or dark ink and light ink) are distributed in a multi-valued stage of about 8 bits, and independent planes are created. As with other colors, a method of performing subsequent image processing is useful. With such a method, an arbitrary cyan density can be divided into large dots and small dots in a state of 8 bits (256 gradations). Appropriate and smooth gradation expression is possible in all gradations.

  However, when planes with different colors and planes with different dot diameters (or densities) are mixed in this way, similar to the above-mentioned image adverse effect caused by the superposition of cyan and magenta, it is caused by the superposition of various elements. There are concerns about evil. That is, as in the prior art, simply focusing on the superposition of cyan and magenta is not sufficient in terms of dot dispersibility.

  In this case, an N-dimensional LUT corresponding to a plurality of planes may be prepared in a configuration as disclosed in Patent Document 4, but the ink color and dot diameter (density) are different in nature. It is an element, and the meaning and importance of separating each other are also different. Therefore, it is not appropriate to separate these N components at the same level.

  The present invention has been made to solve the above-described problems, and its object is to quantize a plurality of types of multilevel signals having two or more types of elements such as “dot diameter” and “color”. The present invention provides an image processing method and a recording system that can disperse and arrange each of a plurality of types of dots in a suitable state when a predetermined image is expressed by pseudo halftone processing.

  Therefore, in the present invention, there is provided an image processing method for processing image data for forming an image on a recording medium using at least dots recorded in k different forms for each of m types of ink, For each of the ink types and the dot forms, the step of obtaining m × k types of multilevel signal values, and for each of the m × k types of multilevel signal values, m × By adding the k types of quantization error values to obtain m × k types of second multilevel signal values, and adding the m × k types of second multilevel signal values to the ink types. , Obtaining m addition values, obtaining m provisional quantized values based on the m addition values by referring to an m-dimensional lookup table, and the m Each provisional quantized value This is separated into k to obtain m × k types of quantized values, and m generated at the target pixel based on the quantized values and the value of the second multilevel signal. And a calculation step of calculating k quantization error values.

  An image processing method for processing image data for forming an image on a recording medium using at least dots recorded in different forms of k stages for each of m types of inks, A step of acquiring m × k types of multilevel signal values for each type and for each dot form, and for each of the m × k types of multilevel signal values, each of the m × k types of quantum generated in neighboring pixels A step of obtaining m × k types of second multi-level signal values by adding the error values, and adding the m × k types of second multi-level signals in k stages to obtain k additional values , Obtaining k temporary quantized values based on the k added values by referring to a k-dimensional lookup table, and the k temporary quantum To separate each digitized value into m Therefore, a separation step for obtaining m × k types of quantization values, and m × k quantization error values generated in the pixel of interest are calculated based on the quantization values and the second multilevel signal value. And a calculating step.

  An image processing method for processing image data including a plurality of signals relating to cyan and a plurality of signals relating to magenta, wherein each cyan signal generated in a neighboring pixel is obtained for each of the plurality of signals relating to cyan included in a target pixel. A cyan error addition step of adding each quantization error value and obtaining an error addition value for each of the plurality of signals, and each of the plurality of signals relating to the magenta included in the target pixel, each generated in a neighboring pixel A magenta error addition step for adding the quantization error values of each of the magenta signals and obtaining an error addition value for each of the plurality of signals, and a cyan signal addition for summing up the respective error addition values obtained in the cyan error addition step Sum of the error addition values obtained in the process and the magenta error addition process. A provisional quantization of cyan by referring to a look-up table based on a magenta signal addition step, a cyan addition value obtained in the cyan signal addition step, and a magenta addition value obtained in the magenta signal addition step; A determination step for determining a provisional quantization value of the value and magenta, and the cyan value determined in the determination step according to an error addition value for each of the plurality of signals obtained in the cyan error addition step A provisional quantized value is distributed to a plurality of cyan signals, and a cyan signal distributing step for determining a quantized value for each of the plurality of cyan signals, and the plurality of magenta error adding steps obtained in the magenta error adding step. The magenta determined in the determining step according to an error addition value for each signal A magenta signal distribution step of distributing a provisional quantized value to the plurality of magenta signals, and determining a quantized value for each of the plurality of magenta signals, the cyan signal distribution step, and the magenta signal distribution step Based on the quantized value of each of the cyan and magenta signals determined in step, and the error added value of each of the cyan and magenta signals obtained in the cyan error adding step and the magenta error adding step. And an error calculation step of calculating a quantization error value generated in the pixel of interest for each of the cyan and magenta signals.

  Further, on the recording medium, at least one of cyan first dots and second dots having different dot diameters or dot densities and magenta first dots and second dots having different dot diameters or dot densities are used. An image processing method for performing data processing on pixel data corresponding to each pixel constituting the image so as to form an image on the basis of pixel data corresponding to the pixel of interest. A first generation step of generating a value signal value, a cyan second dot multi-value signal value, a magenta first dot multi-value signal value, and a magenta second dot multi-value signal value; and the first generation step. A second generation step of generating a cyan dot multi-value signal value by adding the cyan first dot multi-value signal value and the second dot multi-value signal value generated in step S; A third generation step of generating a magenta dot multi-value signal value by adding the magenta first dot multi-value signal value and the second dot multi-value signal value generated in one generation step; and the second In addition, a temporary quantization value of cyan dots and magenta dots for the target pixel is generated based on the cyan dot multi-value signal value and the magenta dot multi-value signal value generated in the third generation step. Based on the generation step, the cyan first dot multilevel signal value, the cyan second dot multilevel signal value, and the provisional quantized value of the cyan dot generated in the fourth generation step A fifth generation step of generating a quantized value of the cyan first dot and a quantized value of the cyan second dot, a multi-value signal value for the first dot of the magenta, and for the second dot of the magenta Multi-value Based on the sign value and the provisional quantized value of the magenta dot generated in the fourth generating step, the quantized value of the first magenta dot and the quantized value of the second magenta dot are generated. And a sixth generation step.

  Further, on the recording medium, at least one of cyan first dots and second dots having different dot diameters or dot densities and magenta first dots and second dots having different dot diameters or dot densities are used. An image processing method for performing data processing on pixel data corresponding to each pixel constituting the image so as to form an image on the basis of pixel data corresponding to the pixel of interest, and a multivalue for cyan first dot A first generation step of generating a signal value, a multi-value signal value for cyan second dot, a multi-value signal value for magenta first dot, and a multi-value signal value for magenta second dot; and the first generation step. A second generation step of adding the generated cyan first dot multi-value signal value and the magenta first dot multi-value signal value to generate a first dot multi-value signal value A second dot multi-value signal value is generated by adding the cyan second dot multi-value signal value generated in the first generation step and the magenta second dot multi-value signal value. Based on the multi-value signal value for the first dot and the multi-value signal value for the second dot generated in the generation step and the second and third generation steps, the first dot and the second dot of the pixel of interest A fourth generation step for generating a provisional quantized value; a multi-value signal value for the first dot of cyan; a multi-value signal value for the first dot of magenta; and the fourth value generated in the fourth generation step. A fifth generation step for generating a quantized value of the first dot of cyan and a quantized value of the first dot of magenta based on a provisional quantized value of one dot; and for the second dot of cyan Multilevel signal value, for the second dot of the magenta Based on the value signal value and the provisional quantized value of the second dot generated in the fourth generating step, the quantized value of the cyan second dot and the quantized value of the magenta second dot are obtained. And a sixth generation step of generating.

  Furthermore, there is provided a recording system for forming an image on a recording medium using m or more inks including m colors capable of recording dots in a k-stage form, and there are m × k types of m colors and k stages. Means for receiving a density signal of values; means for obtaining m × k types of quantized multilevel signal values by adding a quantization error value to each of the m × k types of multilevel signals; The m × k kinds of quantized multilevel signals are added for each ink color to obtain m addition values, and the m addition values are referred to by referring to an m-dimensional lookup table. Means for obtaining m provisional quantized values, separation means for obtaining m × k types of final quantization values by separating each of the m provisional quantum values into k pieces, and the final M × k error values are obtained from the quantum value and the value of the quantized multilevel signal. And means, based on said error value, characterized by comprising means for calculating the quantization error.

  Furthermore, there is provided a recording system for forming an image on a recording medium using inks of m colors or more including m colors capable of recording dots in k stages, and there are m × k types of m colors and k stages. Means for receiving a multi-value density signal, and means for obtaining m × k kinds of quantized multi-value signal values by adding a quantization error value to each of the m × k kinds of multi-value signals. Means for adding the m × k kinds of quantized multi-level signals by k stages to obtain k added values, and by referring to the k-dimensional lookup table, the k additions. Means for obtaining k provisional quantized values from the values, separating means for obtaining m × k types of final quantized values by separating each of the k provisional quantum values into m pieces, and Acquire mxk error values from the final quantum value and the value of the quantized multilevel signal. Means that, based on said error value, characterized by comprising means for calculating the quantization error.

  According to the present invention, even in an output image obtained by superimposing a plurality of planes in which a plurality of elements are mixed, each of a plurality of types of dots can be dispersed and arranged in a suitable state. It is possible to output various images.

(First embodiment)
Hereinafter, a first embodiment of the present invention will be described in detail with reference to the drawings. The recording system applied in this embodiment uses cyan, magenta, black, and yellow inks, and ink jet recording in which cyan and magenta inks can be formed in two levels of large dots and small dots, respectively. System. In the present embodiment, the separation (dispersion) operation unique to the present invention is not performed for the black ink. This is because even if the black dot overlaps with other color dots, the color appears dominant and is hardly affected by the other colors. Also, regarding yellow dots, the visibility is very weak compared to other colors, so separation (dispersion) work is not performed.

  FIG. 1 is a block diagram for explaining the flow of image data conversion processing in the present embodiment. As shown in the figure, each processing shown here is constituted by a recording device and a personal computer (PC) as a host device.

  There are an application and a printer driver as programs that operate in the operating system of the host device, and the application J0001 executes processing for creating image data to be recorded by the recording device. At the time of actual recording, image data created by the application is passed to the printer driver.

  In the recording system of this embodiment, the user can select the recording mode from the printer driver according to the application. For example, whether the image quality mode or the high-speed mode is selected, and the type of the recording medium can be set here. Each process after the printer driver is designed to some extent independently according to the recording mode.

  Assume that the printer driver in this embodiment includes a pre-stage process J0002, a post-stage process J0003, a secondary component division process J0004, a halftoning J0005, and a print data creation J0006. Here, each process will be briefly described. The pre-stage process J0002 performs color gamut mapping. Then, data conversion is performed to map the color gamut reproduced by the image data R, G, B such as the sRGB standard into the color gamut reproduced by the recording apparatus. Specifically, data in which R, G, and B are each expressed in 8 bits is converted into 8-bit data of R, G, and B having different contents by using a three-dimensional LUT.

  The post-stage process J0003 is a process for obtaining the color separation data Y, M, C, and K corresponding to the combination of inks that reproduces the color represented by the data based on the data R, G, and B on which the color gamut is mapped. Do. Here, similarly to the pre-processing, it is assumed that interpolation calculation is performed in combination with a three-dimensional LUT.

In the secondary component division processing J0004, a primary component corresponding to each large dot and a secondary component corresponding to a small dot are obtained from the cyan component (signal) C and the magenta component (signal) M. For example, in order to generate the cyan first component C1 and the second component C2 from the cyan density component C, a one-dimensional lookup table is used. That is,
C1 = LUTC1 (C)
C2 = LUTC2 (C)
M1 = LUTM1 (M)
It can be expressed as M2 = LUTM2 (M).

  For black K and yellow, the secondary component division processing J0004 is output as-is.

Although an independent one-dimensional lookup table is used here for each component, a multidimensional LUT in which the components are related to each other may be used. For example,
(C1, C2, M1, M2) = LUTColor (C, M) may be used. Also consider yellow and black,
(C1, C2, M1, M2, Y, K) = LUTColor (C, M, Y, K) may be used. In any case, the image data after the secondary component division processing J0004 of the present embodiment is a multi-value density in which 6 components of C1, C2, M1, M2, Y, and K are each represented by 8 bits. It is data.

  Halftoning J0005 performs quantization processing for converting each of 8-bit color separation data Y, M1, M2, C1, C2, and K into 2-bit data. In the present embodiment, for K (black) and Y (yellow), 256-level 8-bit data is converted into 3-level 2-bit data using the above-described multilevel error diffusion method.

  FIG. 2 is a flowchart for explaining a process of performing quantization processing in the halftoning J0005 while separating cyan and magenta of the target pixel from each other.

  In step S1000, first, the density value of each component for the pixel of interest P (x, y) is acquired. That is, the cyan large dot component InC1 (x, y), the cyan small dot component InC2 (x, y), the magenta large dot component InM1 (x, y), and the magenta small dot component InM2 (x, y). Get each. The value acquired here is an input value output from the secondary component division process and input to the halftone processing unit J0005.

Next, in steps S1110 and S1130, the error from the surrounding pixels is added to the cyan first component InC1 and the magenta first component InM1, and the quantized multilevel signal C1 (x, y), Obtain M1 (x, y). For example, if the peripheral error of cyan is ErrorC1 (x, y),
C1 (x, y) = InC1 (x, y) + ErrorC1 (x, y). Similarly,
M1 (x, y) = InM1 (x, y) + ErrorM1 (x, y).

Next, in steps S1120 and S1140, the error from the surrounding pixels is added to the second cyan component C2 and the second magenta component M2. That is,
C2 (x, y) = InC2 (x, y) + ErrorC2 (x, y)
M2 (x, y) = InM2 (x, y) + ErrorM2 (x, y).

  The processes in steps S1110, S1120, S1130, and S1140 described above may be performed sequentially or in parallel.

In steps S1210 and S1220, the first component and the second component of each color of cyan and magenta are added. That is, if the added value is C and M, C = C1 (x, y) + C2 (x, y)
M = M1 (x, y) + M2 (x, y).

  In step S1300, OutC and OutM are acquired from C and M obtained by referring to a two-dimensional LUT prepared in advance. Here, OutC and OutM are density values represented by integers of 0 to 6.

  FIG. 3 is a schematic diagram for explaining the two-dimensional LUT referred to in step S1300. In the figure, the horizontal axis indicates C and the vertical axis indicates M. Although the value output from the secondary component division process is represented by 0 to 255, after adding the error value in steps S1110 to S1140, there is a possibility that the density value has a wider area. Therefore, in the LUT of the present embodiment, the added values C and M correspond to values of −85 to 595. If C or M is less than -85, this value is -85, and if it is 595 or more, this value is 595. Output values (OutC, OutM) corresponding to the regions are shown in regions surrounded by a number of oblique lines drawn obliquely. The output values (OutC, OutM) are provisional output values (provisional quantized values) that are considered together with the first and second components of each color.

  By determining OutC and OutM using the mapped LUT as shown in FIG. 3, cyan dots and magenta dots can be suitably dispersed (separated). The reason will be briefly explained. When each color plane is overlapped after quantization for each plane, the overlap of cyan dots and magenta dots occurs unexpectedly depending on the probability. This deteriorated the uniformity of the image and lowered the quality. However, according to the present embodiment, control is performed so that cyan and magenta are not recorded in the same pixel as much as possible, particularly at a low density. Therefore, a favorable state for dispersibility can be obtained. Further, there is an advantage that the ratio of the mixed color of cyan and magenta in the target pixel is stored without depending on the probability even in the middle density to the high density.

Next, in steps S1401 to S1505, OutC and OutM are distributed to the first cyan component OutC1, the second cyan component OutC2, the first magenta component OutM1, and the second magenta component OutM2, and are finally quantized. Get the value.
The distribution method will be described below using cyan as an example. First, in step S1401, it is determined whether or not the addition values C and C1 are greater than zero. If C and C1 are larger than 0, the process proceeds to step S1403, and the first cyan component OutC1 is set to OutC1 = INT (OutC × C1 / C).
It is calculated by the following formula. That is, the internal division calculation is performed so that C1 which is multi-value information of 256 gradations is associated with the density values represented by 0 to 6.

  On the other hand, if C and C1 are 0 or less in step S1401, the process proceeds to step S1402. In step S1402, it is defined that OutC1 = 0.

  In the subsequent step S1404, a cyan second component OutC2 is obtained. Here, it is calculated as OutC2 = OutC−OutC1. However, in this embodiment, the signal value given to OutC1 and OutC2 is in the range of 0-3. Therefore, if the value of OutC1 or OutC2 is greater than 3, these are forcibly converted to 3 in step S1405. The above processing is executed for magenta in steps S1501 to S1505.

  In step S1600, the generated error is calculated for each of the four components.

FIG. 4 is a one-dimensional LUT used for obtaining an error value in step S1600. For example, when OutC1 is 1, LutOut (OutC1) = 85. Since the error value ErrorC1 at the target pixel is the difference between the multivalued C1 obtained in step S1110 and 85,
ErrorC1 = C1−LutOut (OutC1). Similarly,
ErrorC2 = C2-LutOut (OutC2)
ErrorM1 = M1-LutOut (OutM1)
ErrorM2 = M2-LutOut (OutM2).
The LUT shown in FIG. 4 can be applied to all colors, but independent LUTs may be prepared for each color.

In step S1700, the error obtained in step S1600 is distributed to surrounding pixels. In this embodiment, the error generated in the target pixel P (x, y) is transferred to the peripheral pixels P (x + 1, y), P (x-1, y + 1), P (x, y + 1), and P (x + 1, y + 1). scatter. For dispersion, the error is multiplied by weighting factors W0, W1, W2, and W3, and then added to the error value at each pixel position. That is, when the error of the first component of cyan is distributed to surrounding pixels,
ErrorC1 (x + 1, y) = ErrorC1 (x + 1, y) + ErrorC1 × W0
ErrorC1 (x-1, y + 1) = ErrorC1 (x-1, y + 1) + ErrorC1 × W1
ErrorC1 (x, y + 1) = ErrorC1 (x, y + 1) + ErrorC1 × W2
ErrorC1 (x + 1, y + 1) = ErrorC1 (x + 1, y + 1) + ErrorC1 × W3
It becomes. Similar processing is performed in C2, M1, and M2. Of course, the error distribution method is not limited to such a method.

  In step S1800, the process for the pixel of interest ends. Then, the target pixel is moved to the next pixel along the scanning line for processing, and the processing from step S1000 is executed again. That is, when the processing from S1000 to S1800 is expressed as, for example, a function Cal (x, y), the horizontal number of pixels to be processed is Width and the number of vertical pixels is Length, x and y are Width. The function Cal (x, y) is repeated one pixel at a time until the length exceeds the length. The above processing is executed by the halftoning J0005 shown in FIG.

  At the end of the process performed by the printer driver, print data is created by adding print control information to the print image data containing the above-mentioned 2-bit gradation data by the print data creation process J0006.

  The recording apparatus performs a dot arrangement patterning process J0007 on the input print data. The dot arrangement patterning process J0007 of this embodiment will be briefly described below. In the halftone process described above, the number of levels is reduced from 256-level multi-value density information (8-bit data) to 4-level gradation value information (2-bit data). However, information that can be actually recorded by the ink jet recording head of this embodiment is binary information indicating whether or not to record ink. In the dot arrangement patterning process, it plays the role of reducing the multi-value level of 0 to 3 to the binary level that determines the presence or absence of dots. Specifically, in this dot arrangement patterning process J0007, for each pixel represented by 2-bit data of levels 0 to 3, which is an output value from the halftone processing unit, the gradation value (level 0 to 0) of that pixel is represented. A dot arrangement pattern corresponding to 3) is assigned, thereby defining dot on / off for each of a plurality of areas in one pixel, and discharging one bit of “1” or “0” in each area in one pixel. Arrange the data.

  FIG. 5 shows an output pattern example for the input levels 0 to 3 of C1 and C2, which is converted by the dot arrangement patterning process J0007 in the present embodiment. Each level value shown on the left of the drawing corresponds to level 0 to level 3 which are output values from the halftone processing unit. An area composed of 2 vertical areas × 2 horizontal areas arranged on the right side corresponds to an area of one pixel (pixel) output by halftone processing. For example, 600 ppi (pixel / inch; reference value) in both vertical and horizontal directions The size corresponds to the pixel density. Each area in one pixel corresponds to a minimum unit in which dot on / off is defined, and corresponds to a recording density of 1200 dpi (dot / inch; reference value) in both vertical and horizontal directions. The recording apparatus of the present embodiment is designed so that one large ink drop of 2 pl and one small dot of 1 pl are recorded in one area expressed by about 20 μm in both vertical and horizontal directions corresponding to the recording density. ing.

  In the figure, the area filled with a circle indicates the area where each dot is recorded. As the number of levels increases, the number of dots to be recorded increases by one. In the present embodiment, consideration is given to the dot arrangement at each level so that the large dots (C1) and the small dots (C2) do not overlap as much as possible.

  However, such a dot arrangement patterning process does not characterize the present invention. The pattern in the dot arrangement patterning process is not limited to the arrangement shown here, and the effect of the present invention can be obtained even if the dot arrangement patterning process is not applied. Here, only one dot arrangement patterning process is applied as one of the gradation expression methods when the output value from the halftoning J0005 is multilevel in several stages.

  In the present embodiment, the dot arrangement patterning process is performed to determine the dot arrangement to be actually ejected, and this is input as an output signal to the drive circuit J0009 of the recording head J0010.

  In this embodiment, in the superposition of four planes including color elements composed of cyan and magenta and dot diameter elements composed of large dots and small dots, separation (dispersion) is performed for each element in two stages. There is a feature. That is, in the flowchart shown in FIG. 2, among the four planes, C1 and C2 and M1 and M2 are added, respectively, and are roughly divided into only elements of colors C and M. Further, these are preferably dispersed by the LUT shown in FIG. This is the first stage separation (dispersion). Next, the values of large dots and small dots in the same color are obtained from the respective values obtained in the separation (dispersion) at the first stage according to the original multi-value distribution. This is the second stage separation (dispersion).

  In the case of dispersion (separation) in such a two-step process, dots can be formed in the same manner as in the case of considering the superposition of four brains, although it can be realized with a simpler configuration than the case of using a four-dimensional LUT. It is possible to obtain a smooth and uniform image that is suitably dispersed. In addition, a preferable dispersion state between C and M may not be equal to a preferable dispersion state between large dots and small dots. Even in such a case, with the configuration of the present embodiment, it is possible to give characteristics to the first-stage separation method and the second-stage separation method so that each is in a preferable dispersed state. is there.

  In the above description, the inkjet recording system including the host device and the recording apparatus illustrated in FIG. 1 has been described as an example. However, the configuration of the present embodiment is not realized only by such a recording system. For example, all processing performed by the printer driver may be executed by a hardware configuration provided in the recording apparatus.

  FIG. 6 is a block diagram showing an electric circuit configuration for realizing the above-described halftoning process by hardware. In the figure, 1, 2 and 4, 5 are adders for executing Step S1110, Step S1120, Step S1130 and Step S1140 described in FIG. That is, the multi-value signal input to the halftone processing unit J0005, that is, the density value (InC1, InC2, InM1, InM2) output from the secondary component division processing is temporarily stored in the error buffers 18, 19, 20, and 21. Each stored error value is added. The values (C1, C2, M1, M2) obtained here are input to the other adders 3 and 6.

  Reference numerals 3 and 6 denote adders for executing steps S1210 and S1220. That is, the primary component (C1, M1) and the secondary component (C2, M2) are added independently for cyan and magenta.

  Reference numeral 7 denotes output determining means, which stores the LUT described with reference to FIG. The addition values C and M obtained by the adders 3 and 6 are input to the output determination unit 7, and the output determination unit determines OutC and OutM by referring to the stored LUT, and distributes the respective outputs. Output to means 8 and 9.

  In addition to the output value from the output determining means 7, the primary component values (C 1, M 1) and the added values (C, M) of each color are input to the output distribution means 8 and 9. The output distribution means 8 and 9 execute the processing from step S1401 to step S1505 from these three values. That is, the first components OutC1 and OutM1 and the second components OutC2 and OutM2 of each color are determined and output.

  The output values OutC1, OutC2, OutM1 and OutM2 obtained by the output distribution means 8 and 9 are also output to the error calculation means 10, 11, 12 and 13, respectively, in order to execute step S1600. The error calculation means 10, 11, 12, and 13 respectively output the output values OutC1, OutC2, OutM1, and OutM2 and C1, C2, M1, and M2 after the errors are added by the adders 1, 2, 4, and 5, respectively. By comparison, error values ErrorC1, ErrorC2, ErrorM1, and ErrorM2 are calculated and output to error diffusion means 14-17.

  The error diffusion means 14-17 divide the error values ErrorC1, ErrorC2, ErrorM1, and ErrorM2 based on a predetermined weighting coefficient, and store them in the error buffers 18-21 for each address of the peripheral pixels.

  With the configuration of FIG. 6 described above, each process of the flowchart described with reference to FIG. 2 can be realized.

  Note that the second-stage separation (distribution) method of the present embodiment described in steps S1401 to S1505 can be realized by another hardware configuration.

  FIG. 7 is a flowchart for realizing the second stage separation (dispersion) by a method different from the method shown in FIG. In FIG. 7, step S1300 and the previous steps, step S1404, step S1504 and the subsequent steps are the same as those in FIG.

  In this method, first, in step S2401, it is determined whether C1 is greater than zero. If C1 ≦ 0, the process proceeds to step S2402, and the output value OutC1 = 0 of C1 is set. On the other hand, if C> 0, the process advances to step S2403 to determine whether C2 is greater than zero. Here, if C2 ≦ 0, the process proceeds to step S2404, and the output value OutC1 of Out1 is set to OutC. If C2> 0 in step S2403, the process proceeds to step S2405, where OutC1 = INT (OutC × C1 / (C1 + C2)).

  In subsequent step S1404, OutC2 is calculated from OutC2 = OutC-OutC1 using OutC1 obtained in any of steps S2402, S2404, and S2405. If the value of OutC1 or OutC2 is larger than 3, these are forcibly converted to 3 in step S2405.

The above steps are also performed in parallel for magenta. (Steps S2501 to S2505)
FIG. 8 is a block diagram showing an electric circuit configuration for realizing the steps described in FIG. Compared with the case of FIG. 6, the values input to the output distribution means 8 and 9 are not OutC, C1 and C, but OutC, C1 and C2.

  Furthermore, as another separation (dispersion) method at the second stage, instead of using the calculation formula as shown in the above flowchart, it is a method for obtaining all output values at once using a LUT prepared in advance. Also good.

  9 to 11 are diagrams showing examples of LUTs that are referred to in order to perform second-stage separation (dispersion), that is, separation between large dots and small dots. Here, for example, in the case of cyan, a three-dimensional LUT having three parameters OutC, C1, and C2 obtained in step S1300 as input values is obtained.

  FIG. 9 is an LUT for obtaining OutC1 and OutC2 from C1 and C2 when OutC is 0 or 1. In the figure, the horizontal axis is C1 and the vertical axis is C2, and the values of C1 and C2 are uniquely determined for each region surrounded by a solid line.

  FIG. 10 shows the LUT for obtaining OutC1 and OutC2 from C1 and C2 when OutC is 2, as in FIG. Further, FIG. 11 is a diagram showing a case where OutC is 3.

  FIG. 12 is a flowchart showing an example in which the second-stage separation is completed in a one-stage process. Here, for example, the process performed while performing calculation processing in steps S1401 to S1405 in FIG. 2 is completed only in step S3401 by using a three-dimensional LUT. That is, the separation (dispersion) process in the second stage can be executed more quickly.

  A configuration in which a separation process is performed using a two-stage LUT is also disclosed in FIG. However, according to the configuration described in Patent Document 4, the first LUT is a one-dimensional LUT for quantizing the multilevel information of each plane to a lower level, and the second LUT is a cyan LUT. And a multi-dimensional LUT for separating colors such as magenta. On the other hand, the LUT of the present embodiment described above is different from the configuration described in Patent Document 4 because the two LUTs are both multidimensional LUTs for separation. is there. Of course, even if a one-dimensional LUT for further simplifying the LUT is further added to the previous stage of the process in order to obtain the effect of Patent Document 4, it does not affect the characteristics of the present invention.

  As described above, according to the present embodiment, in the inkjet recording system capable of recording cyan ink and magenta ink with large dots and small dots, the signal value (C1) of the same color and the signal value of the small dot (C1) After obtaining the sum (C) with C2), the quantized value (OutC) of each color is provisionally obtained from the sum value (C and M) of each color using a two-dimensional LUT. Then, final quantum values (OutC1 and OutC2) are obtained from the obtained provisional value (OutC) in a state where the colors are not related to each other. As a result, even in an output image obtained by superimposing a plurality of planes, it is possible to obtain uniform quality without graininess.

(Second Embodiment)
The second embodiment of the present invention will be described below. The recording system applied in this embodiment is the same as that in the first embodiment. That is, an ink jet recording system that uses cyan, magenta, black, and yellow inks and that can form cyan and magenta inks in two levels of large dots and small dots, respectively. Also in the present embodiment, it is assumed that the flow of the image data conversion process can be described with reference to FIG. However, the processing method in the halftoning J0005 is different from that in the first embodiment.

  FIG. 13 is a flowchart for explaining the process of quantizing the pixel of interest in this embodiment.

  In step S5000, first, the density value of each component for the pixel of interest P (x, y) is acquired. That is, the cyan large dot component InC1 (x, y), the cyan small dot component InC2 (x, y), the magenta large dot component InM1 (x, y), and the magenta small dot component InM2 (x, y). Get each. The value acquired here is an input value output from the secondary component division process and input to the halftone processing unit J0005.

Next, in steps S5110 and S5130, the error from the surrounding pixels is added to the cyan first component InC1 and second component InC2. For example, if the peripheral error of cyan is ErrorC1 (x, y),
C1 (x, y) = InC1 (x, y) + ErrorC1 (x, y). Similarly,
C2 (x, y) = InC2 (x, y) + ErrorC2 (x, y).

Next, in steps S5120 and S5140, the error from the surrounding pixels is added to the first component M1 and the second component M2 of magenta. That is,
M1 (x, y) = InM1 (x, y) + ErrorM1 (x, y)
M2 (x, y) = InM2 (x, y) + ErrorM2 (x, y).

  The processes in steps S5110, S5120, S5130, and S5140 described above may be performed sequentially or in parallel.

In step S5210, the first components of cyan and magenta are added. That is, if the added value is S1, S1 = C1 (x, y) + M1 (x, y).
Similarly, add the second component of cyan and magenta,
S2 = C2 (x, y) + M2 (x, y).

  In step S5300, OutS1 and OutS2 are acquired from S1 and S2 obtained by referring to a two-dimensional LUT prepared in advance. Although the LUT is not specifically described here, the mapping is such that the large dots and the small dots are distributed in a suitable state, and the output values OutS1 and OutS2 are integers of 0 to 6. Yes.

  Next, in steps S5401 to S5505, OutS1 and OutS2 are distributed to the first cyan component OutC1, the second cyan component OutC2, the first magenta component OutM1, and the second magenta component OutM2.

Hereinafter, S1 (large dot) will be described as an example. First, in step S5401, it is determined whether or not the addition values S1 and C1 are greater than zero. If S1 and C1 are larger than 0, the process proceeds to step S5403, where the first cyan component OutC1 is set to OutC1 = INT (OutS1 × C1 / S1).
It is calculated by the following formula. That is, the internal division calculation is performed so that C1 which is multi-value information of 256 gradations is associated with the density values represented by 0 to 6.

  On the other hand, if S1 is 0 or less in step S5401, the process proceeds to step S5402. In step S5402, it is defined that OutC1 = 0.

  In a succeeding step S5404, a magenta first component OutM1 is obtained. Here, it is calculated as OutM1 = OutS1-OutC1. However, in this embodiment, the signal value given to OutC1 and OutM1 is in the range of 0-3. Therefore, if the value of OutC1 or OutM1 is greater than 3, these are forcibly converted to 3 in step S5405. The above processing is executed for S2 (small dots) in steps S5501 to S5505.

In step S5600, the generated error is calculated for each of the four components. Here, as in the first embodiment, the one-dimensional LUT shown in FIG. 4 can be applied. That is,
ErrorC1 = C1-LutOut (OutC1)
ErrorC2 = C2-LutOut (OutC2)
ErrorM1 = M1-LutOut (OutM1)
ErrorM2 = M2-LutOut (OutM2).

In step S5700, the error obtained in step S5600 is distributed to surrounding pixels in the same manner as in the first embodiment. That is, when the error of the first component of cyan is distributed to surrounding pixels,
ErrorC1 (x + 1, y) = ErrorC1 (x + 1, y) + ErrorC1 × W0
ErrorC1 (x-1, y + 1) = ErrorC1 (x-1, y + 1) + ErrorC1 × W1
ErrorC1 (x, y + 1) = ErrorC1 (x, y + 1) + ErrorC1 × W2
ErrorC1 (x + 1, y + 1) = ErrorC1 (x + 1, y + 1) + ErrorC1 × W3
It becomes. Similar processing is performed in C2, M1, and M2. Of course, the error distribution method is not limited to such a method.

  In step S5800, the process for the pixel of interest ends. Then, the target pixel is moved to the next pixel along the scanning line for processing, and the processing from step S5000 is executed again. That is, when the processing from S5000 to S5800 is expressed as a function Cal (x, y), for example, if the horizontal pixel number to be processed is Width and the vertical pixel number is Length, x and y are Width The function Cal (x, y) is repeated one pixel at a time until the length exceeds the length. The above processing is executed by the halftoning J0005 shown in FIG.

  As described above, according to the present embodiment, in an inkjet recording system capable of recording cyan ink and magenta ink with large dots and small dots, the sum between different colors in the signal value of large dots (S1 = C1 + M1), and After obtaining the sum between the different colors in the signal value of the small dots, the quantized value (OutS1) of each color is provisionally obtained from each sum value (S1) using a two-dimensional LUT. Thereafter, the final quantum value (OutC1, OutM1) is obtained from the obtained provisional value (OutS1) while the dot diameters are not related to each other. As a result, even in an output image obtained by superimposing a plurality of planes, it is possible to obtain uniform quality without graininess.

  As in the present embodiment, the method of performing separation between colors in the second stage after separating the size of the dot diameter in the first stage is different from the first embodiment in the state of dot dispersion. Is expected. In general, a more preferable form may be selected from the color development state of the applied ink, the degree of visibility, and the like. If necessary, the first embodiment and the second embodiment may be switched according to the type of recording medium and the recording mode.

(Other embodiments)
In the first and second embodiments described above, as shown in FIG. 1, the component of the similar color (for example, cyan) is divided into two components (C1, C2) in the secondary generation division processing J0004. In the present invention, it is not essential to provide the secondary generation division processing J0004. For example, the post-processing unit J0003 converts RGB signals into six component signals of C1, C2, M1, M2, Y, and K, and inputs the six component signals to the halftone processing unit J0005. As described in the first and second embodiments, the large and small CM separation processing may be performed.

In the above embodiment, an inkjet recording system having a configuration capable of recording large dots and small dots for specific similar colors (cyan and magenta) has been described as an example. However, the present invention is limited to such a configuration. It is not something. As a method of expressing the density of one pixel in multiple stages while using similar colors, there is a method of preparing inks having different densities in addition to a method of varying the dot diameter. Even in such a case, the present invention is effective. In other words, by replacing the large dots and small dots described above with dark ink and light ink, separation is performed with the same effect for the dark cyan, light cyan, dark magenta, and light magenta planes output in multiple values. Processing can be realized. Further, for similar colors, both the dot diameter and the dot density may be different. As described above, the present invention can be applied to a form in which at least one of the dot density and the dot diameter is different for similar colors.
The effects of the present invention are not obtained only when recording is performed with four colors of cyan, magenta, yellow, and black. There are many types of ink colors and dot sizes, and the ink colors to be separated are not limited to cyan and magenta. For example, in addition to the above four colors, the present invention can also be applied to a configuration in which red, blue, and green can be recorded with large dots, small dots, and medium dots. In this case, for example, if the colors are separated by the first separation process as in the first embodiment, the LUT used here has dimensions corresponding to the number of colors.

  Further, even in halftoning, the fact that the output value is a 4-bit quaternary signal does not characterize the present invention. Halftoning output values (OutC1, OutC2, OutM1, OutM2) may be binary. In this case, the dot arrangement patterning process shown in FIG. 1 is not necessary.

  The output destination after halftoning is not limited to the ink jet recording apparatus as long as it is a recording apparatus that expresses an image with a dot matrix. In addition to the host device shown in FIG. 1, a storage medium (floppy disk (registered trademark)) for storing the multi-valued image other than the host device shown in FIG. , CD-ROM, etc.), reception via a network, image scanner or the like.

  The feature of the present invention lies in a technique for quantizing a plurality of types of multilevel signals composed of two or more types of elements such as colors and dot diameters. This characteristic processing is performed by hardware. It is included in the scope of the present invention whether it is realized by software or by software.

It is a block diagram for demonstrating the flow of the image data conversion process in embodiment of this invention. It is a flowchart for demonstrating the process of quantizing each, separating cyan and magenta of a pixel of interest from each other. It is a schematic diagram which shows the two-dimensional LUT referred by step S1300 of FIG. This is a one-dimensional LUT used to obtain an error value in step S1600. It is a figure which shows the example of an output pattern with respect to the input levels 0-3 of C1 and C2 converted by the dot arrangement patterning process in this embodiment. It is the block diagram which showed the electric circuit structure for implement | achieving a halftoning process. It is a flowchart for implement | achieving isolation | separation (distribution) of a 2nd step by another method. It is the block diagram which showed the electric circuit structure for implement | achieving the process demonstrated in FIG. This is an LUT for obtaining OutC1 and OutC2 from C1 and C2. This is an LUT for obtaining OutC1 and OutC2 from C1 and C2. This is an LUT for obtaining OutC1 and OutC2 from C1 and C2. It is a flowchart of the example which completes separation of the 2nd step in one process. It is a flowchart for demonstrating the process of performing the quantization of the focused pixel in the 2nd Embodiment of this invention.

Explanation of symbols

1 to 6 Adder 7 Output decision means 8, 9 Output distribution means 10 to 13 Error calculation means 14 to 17 Error distribution means 18 to 21 Error buffer

Claims (18)

An image processing method for processing image data for forming an image on a recording medium using at least dots recorded in different forms of k stages for each of m types of ink,
For the pixel of interest, obtaining m × k types of multilevel signal values for each ink type and each dot form;
A step of obtaining m × k types of second multilevel signal values by adding m × k types of quantization error values generated in neighboring pixels to each of the m × k types of multilevel signal values. When,
Adding the m × k types of second multi-value signal values to the ink type to obtain m added values;
obtaining m provisional quantization values based on the m addition values by referring to an m-dimensional lookup table;
Separating each of the m provisional quantized values into k, thereby obtaining m × k kinds of quantized values;
A calculation step of calculating m × k quantization error values generated in the target pixel based on the quantization value and the value of the second multi-level signal;
An image processing method comprising: An image processing method for processing image data for forming an image on a recording medium using at least dots recorded in different forms of k stages for each of m types of ink,
For the pixel of interest, obtaining m × k types of multilevel signal values for each ink type and each dot form;
A step of obtaining m × k types of second multilevel signal values by adding m × k types of quantization error values generated in neighboring pixels to each of the m × k types of multilevel signal values. When,
Adding the m × k types of second multi-value signals by k stages to obtain k added values;
obtaining k provisional quantization values based on the k addition values by referring to a k-dimensional lookup table;
Separating each of the k provisional quantized values into m, thereby obtaining m × k kinds of quantized values;
A calculation step of calculating m × k quantization error values generated in the target pixel based on the quantization value and the second multi-level signal value;
An image processing method comprising:   3. The image processing method according to claim 1, wherein, in the separation step, the provisional quantization value is separated according to the addition value and the second multi-level signal value.   3. The image processing method according to claim 1, wherein, in the separation step, the provisional quantized values are separated by referring to a multidimensional lookup table.   The image processing method according to claim 1, wherein the m kinds of ink include cyan ink and magenta ink.   6. The image processing method according to claim 1, wherein the size of the dots formed on the recording medium is different in k stages from the different forms in the k stages.   The image processing method according to claim 1, wherein the density of the ink is different in k stages from the different forms in the k stages.   The image processing method according to claim 1, wherein the quantization value acquired in the separation step is binary.   The image processing method according to claim 1, wherein the quantized value acquired in the separation step is a multilevel of several stages. An image processing method for processing image data including a plurality of signals relating to cyan and a plurality of signals relating to magenta,
A cyan error addition step of adding a quantization error value of each cyan signal generated in a neighboring pixel for each of the plurality of cyan-related signals included in the pixel of interest and obtaining an error addition value for each of the plurality of signals; ,
A magenta error addition step of adding a quantization error value of each magenta signal generated in a neighboring pixel to each of a plurality of signals relating to the magenta included in the target pixel and obtaining an error addition value for each of the plurality of signals. When,
A cyan signal addition step of summing up the respective error addition values obtained in the cyan error addition step;
A magenta signal addition step of summing up the respective error addition values obtained in the magenta error addition step;
By referring to the lookup table based on the cyan addition value obtained in the cyan signal addition step and the magenta addition value obtained in the magenta signal addition step, the provisional quantization value of cyan and the provisional value of magenta A determination step for determining a quantized value;
In accordance with the error addition value for each of the plurality of signals obtained in the cyan error addition step, the provisional quantized values for cyan determined in the determination step are distributed to the cyan signals. A cyan signal distribution step for determining a quantized value for each of the cyan signals;
According to the error addition value for each of the plurality of signals obtained in the magenta error addition step, the provisional quantized value related to the magenta determined in the determination step is distributed to the plurality of magenta signals. A magenta signal distribution step for determining a quantization value for each of the plurality of signals of the magenta; a quantization value of each of the cyan and magenta signals determined in the cyan signal distribution step and the magenta signal distribution step; Based on the error addition value of each of the cyan and magenta signals obtained in the cyan error addition step and the magenta error addition step, the quantization error value generated in the pixel of interest is converted into a plurality of cyan and magenta signals. An error calculation process to calculate for each signal;
An image processing method comprising: An image is formed on a recording medium by using at least one cyan first dot and second dot having different dot diameters or dot densities, and at least one magenta first dot and second dots having different dot diameters or dot densities. An image processing method for performing data processing on pixel data corresponding to each pixel constituting the image,
Based on the pixel data corresponding to the target pixel, the first multi-value signal value for cyan, the second multi-value signal value for cyan, the first multi-value signal value for magenta, and the second multi-value signal value for magenta. A first generation step of generating a value signal value;
A second generation step of adding the cyan first dot multi-value signal value and the second dot multi-value signal value generated in the first generation step to generate a cyan dot multi-value signal value;
A third generation step of generating a magenta dot multi-value signal value by adding the magenta first dot multi-value signal value and the second dot multi-value signal value generated in the first generation step;
Based on the cyan dot multi-value signal value and the magenta dot multi-value signal value generated in the second and third generation steps, a provisional quantized value of cyan dots and magenta dots for the target pixel is generated. A fourth generation step to
Based on the cyan first dot multilevel signal value, the cyan second dot multilevel signal value, and the provisional quantization value of the cyan dot generated in the fourth generation step, the cyan A fifth generation step of generating a quantization value of the first dot and a quantization value of the cyan second dot;
Based on the magenta first dot multi-value signal value, the magenta second dot multi-value signal value, and the provisional quantization value of the magenta dot generated in the fourth generation step, A sixth generation step of generating a quantized value of the first dot and a quantized value of the second dot of the magenta;
An image processing method comprising:   In the fourth generation step, a two-dimensional lookup table in which the cyan dot multi-value signal value and the magenta dot multi-value signal value are associated with the provisional quantized values of the cyan dot and magenta dot is used. From the cyan dot multi-value signal value generated in the second generation step and the magenta dot multi-value signal value generated in the third generation step, provisional provision of cyan dots and magenta dots for the target pixel is performed. The image processing method according to claim 10, wherein a quantized value is generated. The first generation step includes
(A) The pixel data corresponding to the pixel of interest includes four independent multi-value signal values corresponding to the first cyan dot, the second cyan dot, the first magenta dot, and the second magenta dot, respectively. A conversion process for converting to
(B) For each of the four converted multi-value signal values, the first pixel of cyan, the second dot of cyan, the first dot of magenta, and the second dot of magenta are generated in neighboring pixels. By adding an error, the cyan first dot multi-value signal value, the cyan second dot multi-value signal value, the magenta first dot multi-value signal value, the magenta Generating a second dot multi-value signal value,
The error corresponding to the cyan first dot is the difference between the cyan first dot multi-value signal value generated in the first generation step and the cyan first dot generated in the fifth generation step. Calculated based on the quantized value,
The error corresponding to the cyan second dot is the difference between the cyan second dot multi-value signal value generated in the first generation step and the cyan second dot generated in the fifth generation step. Calculated based on the quantized value,
The error corresponding to the first dot of magenta includes the multi-value signal value for the first dot of magenta generated in the first generation step and the first dot of magenta generated in the sixth generation step. Calculated based on the quantized value,
The error corresponding to the second magenta dot is the difference between the magenta second dot multi-value signal value generated in the first generation step and the second magenta dot generated in the sixth generation step. The image processing method according to claim 10, wherein the image processing method is calculated based on a quantization value. An image is formed on a recording medium by using at least one cyan first dot and second dot having different dot diameters or dot densities, and at least one magenta first dot and second dots having different dot diameters or dot densities. An image processing method for performing data processing on pixel data corresponding to each pixel constituting the image,
Based on the pixel data corresponding to the target pixel, the first multi-value signal value for cyan, the second multi-value signal value for cyan, the first multi-value signal value for magenta, and the second multi-value signal value for magenta. A first generation step of generating a value signal value;
Second generation for generating a first dot multilevel signal value by adding the cyan first dot multilevel signal value generated in the first generation step and the magenta first dot multilevel signal value Process,
Third generation for generating a second dot multi-value signal value by adding the cyan second dot multi-value signal value generated in the first generation step and the magenta second dot multi-value signal value Process,
Based on the first dot multi-value signal value and the second dot multi-value signal value generated in the second and third generation steps, provisional quantum of the first dot and the second dot for the pixel of interest A fourth generation step of generating a conversion value;
Based on the cyan first dot multi-value signal value, the magenta first dot multi-value signal value, and the provisional quantized value of the first dot generated in the fourth generation step, the cyan A fifth generation step of generating a first dot quantization value and a magenta first dot quantization value;
Based on the cyan second dot multilevel signal value, the magenta second dot multilevel signal value, and the provisional quantized value of the second dot generated in the fourth generation step, the cyan A sixth generation step of generating a second dot quantization value and a magenta second dot quantization value;
An image processing method comprising:   The cyan first dot is a dot having a relatively large diameter compared to the cyan second dot, and the magenta first dot is relatively large in diameter compared to the magenta second dot. The image processing method according to claim 11, wherein the image processing method is a dot having a dot.   The first cyan dot is a dot having a relatively high density compared to the second cyan dot, and the first magenta dot has a relatively higher density than the second magenta dot. The image processing method according to claim 10, wherein the image processing method is a dot having a dot. A recording system for forming an image on a recording medium using m or more inks including m colors capable of recording dots in a k-stage form,
means for receiving m × k types of multi-value density signals for each of m colors and k stages;
Means for obtaining m × k types of quantized multilevel signal values by adding quantization error values to each of the m × k types of multilevel signals;
Means for adding m × k types of quantized multilevel signals for each ink color to obtain m added values;
means for obtaining m provisional quantization values from the m addition values by referring to an m-dimensional lookup table;
Separating means for obtaining m × k types of final quantized values by separating each of the m provisional quantum values into k pieces;
Means for obtaining m × k error values from the final quantum value and the value of the quantized multilevel signal;
A recording system comprising: means for calculating the quantization error based on the error value. A recording system for forming an image on a recording medium using m or more inks including m colors capable of recording dots in a k-stage form,
means for receiving m × k types of multi-value density signals for each of m colors and k stages;
Means for obtaining m × k types of quantized multilevel signal values by adding quantization error values to each of the m × k types of multilevel signals;
Means for adding the m × k kinds of quantized multilevel signals for each of k stages to obtain k added values;
means for obtaining k provisional quantization values from the k addition values by referring to a k-dimensional lookup table;
Separating means for obtaining m × k kinds of final quantized values by separating each of the k provisional quantum values into m,
Means for obtaining m × k error values from the final quantum value and the value of the quantized multilevel signal;
A recording system comprising: means for calculating the quantization error based on the error value.

Images