Image processing apparatus, threshold matrix generation apparatus, recording apparatus, image processing method, threshold matrix generation method, and program

Abstract

To further equalize the frequency of use of nozzles in recording an image with a plurality of nozzle rows.SOLUTION: An image processing apparatus generates data for forming an image represented by input image data on a recording medium by using a recording head having nozzle rows arranged in plurality (L rows) in a second direction and each having a plurality of nozzles arranged in a first direction. The image processing apparatus converts the input image data into a first dot pattern representing dots recorded by the recording head based on a threshold matrix, assigns the dots in the first dot pattern to each of the nozzle rows so that the nozzles of the recording head are used in a period of the L rows, and obtains a second dot pattern representing recording dots for every nozzle row. The power spectrum of a frequency space in the first dot pattern has blue noise characteristics or green noise characteristics in which the power in a low-frequency band is reduced, and the powers of a frequency corresponding to the period of the L rows and its multiple frequency in the second direction are reduced.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to an image processing device, a threshold matrix generating device, and a method and program thereof. [Background technology]

[0002] Inkjet recording devices are generally known as computer output devices. Inkjet recording devices have a recording head with an array of ink ejection ports (nozzles), and form a desired image on the recording medium by moving the recording head relative to the recording medium and ejecting ink droplets (dots) from the nozzles. In particular, inkjet printers for commercial printing are often used to perform single-pass drawing (also called full-line or full-multi) because of the need for high print productivity. In single-pass drawing, a long line head is used that has a nozzle array that covers the entire range of the drawing area in the width direction of the recording medium, which is perpendicular to the direction in which the recording medium is transported. Hereinafter, the direction in which the recording medium is transported is referred to as the "transport direction" or "X direction." Also, the width direction of the recording medium is referred to as the "recording medium width direction" or "Y direction." Since the recording medium only needs to be moved relative to the line head once, the single-pass drawing method has the characteristic of being faster in printing speed than the multi-pass method, which completes an image by multiple scans.

[0003] When printing with such a printer, image processing is performed to express the gradation values ​​of the original image data to be printed by turning dots on and off, a process known as halftone processing. The printer executes the print process according to the image data after this halftone processing. Various methods have been proposed for halftone processing, one of the most representative methods being the dither method. In the dither method, a dither matrix of a given size, in which different threshold values ​​are arranged, is repeatedly deployed in a tile-like manner on the image data, and the gradation values ​​(pixel values) of the input image are compared to the corresponding threshold values. If the gradation value is greater than the threshold value, the dots are turned on, and if the gradation value is less than the threshold value, the dots are turned off to express the gradation.

[0004] In the dither method, the dot pattern generated by the dither matrix may have a bias in the number of ejection dots (ejection frequency) for each nozzle. Nozzles with a high ejection frequency have a short lifespan, shortening the lifespan of the head itself. Patent Document 1 discloses a technology that suppresses the bias in the number of ejection dots for each nozzle by using a dither matrix that makes the number of ejection dots for each nozzle uniform. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2020-015303 A Summary of the Invention [Problem to be solved by the invention]

[0006] In recent years, a printing device has been developed that has multiple nozzle arrays in the X direction that cover the entire range of the drawing area in the Y direction for one color of ink. For example, when printing with eight nozzle arrays for one color of ink, even if the driving frequency of each nozzle array is the same as when printing with one nozzle array, the scanning speed of the nozzle array can be increased by eight times, and the printing speed can be increased by eight times. However, Patent Document 1 does not consider the number of ejections for each nozzle array when assigning printing dots to such multiple nozzle arrays. Therefore, when a dot pattern generated using the dither matrix of Patent Document 1 is assigned to multiple nozzle arrays, the number of ejection dots for each nozzle array is not uniform.

[0007] Also, by preparing eight dither matrices for each nozzle row, dividing the pre-quantization multi-value data into eight equal parts, and quantizing them with each dither matrix, it is possible to generate a dot pattern with a uniform number of ejections for each nozzle row. However, a dot pattern obtained by adding up the dot patterns for each nozzle row does not take into account the dispersion of the dot arrangement, which can lead to degradation of image quality, such as deterioration of graininess.

[0008] One aspect of the present invention provides a technique for more uniformly using the nozzles when printing an image with a plurality of nozzle rows. [Means for solving the problem]

[0009] An image processing device according to an aspect of the present invention includes the following arrangement: An image processing device that generates data for forming an image corresponding to multi-value input image data on a recording medium using a recording head in which a nozzle row, each having a plurality of nozzles arranged in a first direction, is arranged in L rows (L is 2 or more) in a second direction perpendicular to the first direction, the image processing device comprising: a conversion means for converting the input image data into a first dot pattern representing dots to be printed by the print head based on a threshold matrix in which threshold values ​​are arranged; an allocation means for allocating dots of the first dot pattern to each of the nozzle arrays so that each nozzle of the recording head is used at a period of the L arrays, and obtaining a second dot pattern representing dots recorded by each of the nozzle arrays; The power spectrum in frequency space of the first dot pattern has blue noise or green noise characteristics with power suppressed in the low frequency band, and power is suppressed at frequencies corresponding to the period of the L columns in the second direction and at multiples of the frequencies. Effect of the Invention

[0010] According to the present invention, when an image is printed using a plurality of nozzle rows, the frequency with which each nozzle is used is made more uniform. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram illustrating an inkjet printer according to an embodiment. [Diagram 2] FIG. 2 is a diagram illustrating a recording head according to the embodiment. [Diagram 3] FIG. 1 is a block diagram showing an example of the functional arrangement of a printing system according to an embodiment. [Figure 4] 4 is a flowchart showing image processing according to an embodiment. [Diagram 5] 5A to 5C are views for explaining a nozzle allocation process according to an embodiment. [Figure 6] 5A to 5C are diagrams showing examples of pseudo-halftone data applicable to the embodiment. [Figure 7] Examples of nozzle division patterns applicable to the embodiment [Figure 8] 6 is a flowchart showing a nozzle allocation process according to the embodiment. [Figure 9] 6A to 6C are views for explaining the results of nozzle allocation processing according to the embodiment. [Figure 10] 10A to 10C are diagrams showing examples of nozzle division patterns that can be applied to modified examples of the embodiment. [Figure 11] 10A to 10C are diagrams showing examples of pseudo-halftone data applicable to a modified example of an embodiment. [Figure 12] 10A and 10B are diagrams showing the results of a nozzle allocation process according to a modified example of the embodiment. [Figure 13] 10A and 10B are diagrams for explaining the configuration of a nozzle allocation process according to another embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, the embodiments will be described in detail with reference to the attached drawings. Note that the following embodiments do not limit the invention according to the claims. Although the embodiments describe a number of features, not all of these features are essential to the invention, and the features may be combined in any manner. Furthermore, in the attached drawings, the same reference numbers are used for the same or similar configurations, and duplicated descriptions are omitted.

[0013] FIG. 1 is a diagram showing a schematic diagram of an inkjet printer (inkjet recording device) according to an embodiment. As shown in FIG. 1, the printer 4 includes recording heads 101 to 104 arranged on a frame that constitutes the structure of the printer. The recording heads 101 to 104 are so-called full-line type recording heads, and each of them has a plurality of nozzles for ejecting a plurality of inks of black, cyan, magenta, and yellow arranged along a predetermined direction in a range corresponding to the width of a recording paper 106. The resolution of the nozzle arrangement in each nozzle row is, for example, 1200 dpi. Hereinafter, black, cyan, magenta, and yellow may be expressed as B, C, M, and Y, respectively. The recording paper 106 as a recording medium is transported in the direction of the arrow (X direction) in the figure by the transport roller 105 (and other rollers not shown) rotating by the driving force of a motor (not shown). By the ink ejection operation repeatedly performed by the recording heads 101 to 104 on the transported recording paper 106, for example, an image for one sheet of the recording paper 106 can be recorded.

[0014] As shown in FIG. 2, each of the recording heads 101 to 104 has a configuration in which a nozzle array having a plurality of nozzles for recording ink droplets in the recording medium width direction (Y direction) is arranged in L arrays (L is 2 or more) in the transport direction (X direction) perpendicular to the recording medium width direction. Each of the recording heads 101 to 104 ejects ink onto the recording medium moving relatively in the direction indicated by the arrow (X direction), and performs image recording by dividing the recording dots of the same line (line in the X direction) on the recording medium among the plurality of nozzle arrays. In this embodiment, the number of nozzle arrays L is 8 (L1, L2, L3, L4, L5, L6, L7, L8). Discharge energy generating means such as a heater is provided for each nozzle of each nozzle array constituting each of the recording heads 101 to 104, and power for driving these heaters is supplied via different wiring for each nozzle array. Focusing on a target recording line 107, which is one of the recording lines of a certain color component, recording at dot positions on the target recording line 107 is performed by ejection or non-ejection of nozzles belonging to any of the nozzle rows (L1 to L8).

[0015] (Device configuration) FIG. 3 is a block diagram showing the configuration of an image forming system according to this embodiment. In FIG. 3, the image forming system has an image processing device 3 and a printer 4. The image processing device 3 can be implemented by a printer driver installed in an information processing device such as a general personal computer. In this case, each functional unit of the image processing device 3 described below is realized by a computer (processor) executing a predetermined program. In addition, as another configuration, for example, the printer 4 may include at least a part of the functional units of the image processing device 3 shown in FIG. 3. The image processing device 3 and the printer 4 are connected by a printer interface or a circuit. The image processing device 3 acquires image data to be printed from an image data input terminal 301. The image data is, for example, an 8-bit RGB color image.

[0016] The color matching processing unit 302 performs color matching processing on the input RGB image data and corrects the colors of the RGB image data. The color matching processing makes it possible to obtain uniform color reproduction even when a printer or recording medium having different color reproduction characteristics is used. In the color matching processing, the color matching processing unit 302 refers to a three-dimensional color matching LUT stored in the color matching LUT storage unit 303. In the color matching LUT, for example, RGB values ​​are described on lattice points thinned to 17×17×17 points. The color matching processing unit 302 calculates values ​​between lattice points by linear interpolation of the color matching LUT.

[0017] The color separation processing unit 304 generates a four-plane 8-bit ink value image corresponding to the four colors of ink provided in the printer 2 from the image data corrected by the color matching processing unit 302. In this embodiment, the print heads 101 to 104 print images using four colors of ink: cyan (C), magenta (M), yellow (Y), and black (K). In the color separation processing, a three-dimensional color separation LUT stored in the color separation LUT storage unit 305 is referenced. In the color separation LUT, for example, color material amount values ​​(ink values) of four colors of ink are described on lattice points thinned to 17×17×17 points. The color separation processing unit 304 calculates values ​​between lattice points by linear interpolation of the color separation LUT.

[0018] The quantization processing unit 306 converts the multi-value data (ink value image) corresponding to each color obtained by the color separation processing unit 304 into binary or more than binary image data (dot pattern) with a number of gradations less than the number of input gradations. In this embodiment, the quantization processing unit 306 performs quantization by comparing the dither matrix for each ink color stored in the dither matrix storage unit 307 with the ink value image. The dither matrix is ​​an example of a threshold matrix in which thresholds are arranged to have a predetermined characteristic, and can have various sizes and shapes such as 512×512 pixels, 256×256 pixels, and 256×512 pixels. In this embodiment, a 64×64 pixel dither matrix in which thresholds from 0 to 4095 are arranged is used as an example of the threshold matrix. The dither matrix is ​​repeatedly arranged in a tile shape for the entire image data. As a result, any of the thresholds of the dither matrix is ​​assigned to all pixels. The quantization processing unit 306 compares each pixel value of the ink value image of each color with the corresponding threshold value of the dither matrix of the corresponding ink color to determine whether to print (1) or not print (0) a dot at each pixel position (dot position) and generate a dot pattern. The threshold matrix generating device 5 has a processor (e.g., a CPU) and memory (not shown), and realizes various processes by the processor executing a program stored in the memory. The threshold matrix generating device 5 generates a dither matrix for each ink color and stores it in the dither matrix storage unit 107. The threshold matrix generating device 5 may be configured separately from the image processing device 3, or may be incorporated as a part of the functions executed by the image processing device 3. A method for generating a threshold matrix (a dither matrix in this example) by the threshold matrix generating device 5 will be described later.

[0019] The nozzle allocation processing unit 308 divides the image data (dot pattern) quantized by the quantization processing unit 306 based on the nozzle allocation pattern corresponding to each nozzle array (L1 to L8) of the print heads 101 to 104. The nozzle allocation pattern is stored in the allocation pattern storage unit 309. The nozzle allocation pattern in this embodiment is a binary pattern in which values ​​1 and 0 are arranged as information on whether or not a dot can be allocated. The nozzle allocation processing unit 308 obtains the dot pattern to be printed by each nozzle array (L1 to L8) by performing a logical product between the allocation pattern and the image data (dot pattern). In order to make the nozzle allocation pattern correspond to all pixels of the image data, the nozzle allocation pattern is repeatedly arranged in a tile-like manner for the entire image data (dot pattern). Note that the size Tx in the X direction of the nozzle allocation pattern is desirably an integer multiple of the number of nozzle arrays so that dots are evenly allocated to the nozzle arrays (L1 to L8). In addition, in order not to interfere with the size of the dither matrix when repeatedly arranged in a tile-like manner, the size Tx in the X direction is desirably a divisor of the size Sx in the X direction of the dither matrix. Alternatively, the size Sx of the dither matrix in the X direction may be a multiple of the size Tx of the nozzle allocation pattern in the X direction. Various sizes can be applied to the size Ty of the nozzle allocation pattern in the Y direction. The dot pattern for each nozzle row generated by the nozzle allocation processing unit 308 is output to the printer 4 from the output terminal 310.

[0020] The printer 4 forms a dot pattern generated by the image processing device 3 on a recording medium. In this embodiment, the recording heads 101, 102, 103, and 104 of an inkjet type or the like shown in FIG. 1 are used as the recording heads. The recording heads 101 to 104 are long line heads in which a plurality of nozzles for ejecting ink are arranged to cover the entire range of a drawing area in the width direction of the recording medium, and a printed image is formed by ejecting ink while moving the recording medium relatively. In this embodiment, the recording heads 101, 102, 103, and 104 are equipped with four colors of ink: cyan (C), magenta (M), yellow (Y), and black (K). The input terminal 401 receives image data indicating a dot pattern that has been quantized from the image processing device 3 and assigned to each nozzle row, and passes it to an ink color selection unit 402.

[0021] An ink color selection unit 402 selects an ink color that corresponds to the dot pattern for each nozzle row from among the ink colors mounted on the print heads 101, 102, 103, and 104. A head drive unit 403 generates drive signals for controlling the print heads 101, 102, 103, and 104 to print the dot pattern for each nozzle row with the selected ink color. Based on the drive signals, the print heads 101, 102, 103, and 104 actually print each ink dot on the print medium.

[0022] 3 does not limit which processes are performed by the image processing device 3 and which processes thereafter are performed by the printer 4. For example, if the image processing device 3 performs processes up to quantization, the quantized dot pattern is transferred to the printer 4, and the nozzle allocation process is performed by the printer 4. Depending on the performance of the printer 4, it may be possible to directly receive multi-value RGB image data and perform all processes from the color matching process unit 302 to the nozzle allocation process unit 308 within the printer 4.

[0023] (Image processing flow) FIG. 4 is a flow chart for explaining the processing of image data in the image processing device 3 of FIG. 3. First, the image processing device 3 acquires an input image from the image data input terminal 301 (S400). The color matching processing unit 102 performs color matching processing on the input image (image data) acquired via the image data input terminal 301 (S401). In the color matching processing, a three-dimensional color matching LUT stored in the color matching LUT storage unit 103 is referenced. The color separation processing unit 304 generates color-separated ink value image data from the image data that has been color-matched by the color matching processing unit 102 (S402). In the color separation processing, a three-dimensional color separation LUT stored in the color separation LUT storage unit 305 is referenced. The quantization processing unit 306 acquires the ink value image data after color separation, and performs quantization processing to convert the image data into a binary dot pattern (S403). In the quantization processing, a two-dimensional dither matrix stored in the dither matrix storage unit 307 is referenced.

[0024] Next, the nozzle allocation processing unit 308 performs a nozzle allocation process to divide the dot pattern after the quantization process into dot patterns for each nozzle row based on the nozzle allocation pattern stored in the allocation pattern storage unit 309 (S404). FIG. 5 shows three examples of nozzle allocation patterns for each nozzle row (L1 to L8). The nozzle allocation pattern shown in FIG. 5 has a size of 64×64, with white pixels (pixel value=1) indicating positions to which dots can be allocated and black pixels (pixel value=0) indicating positions to which dots cannot be allocated. Allocation pattern 1 in FIG. 5 is a pattern in which allocation is performed such that dots at the same position in the X direction are ejected consecutively in the Y direction by the same nozzle row. Allocation pattern 2 is a pattern in which allocation is performed such that dots at the same position in the X direction are not ejected consecutively in the Y direction by the same nozzle row. Allocation pattern 3 is a pattern in which irregular allocation is performed. For each allocation pattern, allocation can be made in a period of 8 pixels, which is the number of nozzle rows (L1 to L8), and by shifting the same pattern by one pixel in the X direction for each nozzle row (L1 to L8), the pixels that can be allocated to each nozzle row are exclusive.

[0025] 4, the dot pattern for each nozzle array after the nozzle allocation process is output from the output terminal 310 in any size, such as the entire image or the band width for each unit printing area (S405). This completes a series of image processes by the image processing device 3.

[0026] (Nozzle allocation process details) The inventor of the present application found that the number of dots discharged from each nozzle array is not uniform when a nozzle allocation process for printing with a plurality of nozzle arrays is performed on a dot pattern generated by a conventional quantization process. This problem will be explained with reference to Figs. 6 and 7. Fig. 6(a) shows an example of a dot pattern before nozzle allocation, which is generated by performing a quantization process using a 64 x 64 dither matrix created based on the conventional technology. Fig. 6(b) shows the result of acquiring a dot pattern for each nozzle array in which nozzles are allocated according to nozzle allocation patterns 1 to 3 shown in Fig. 5. Fig. 7 shows the result of counting the number of dots discharged by each of a plurality of nozzles arranged in the Y direction (the number of dots in the X direction) for the dot pattern shown in Fig. 6(b). The horizontal axis of each graph shows the nozzle position (Y direction position), and the vertical axis shows the number of dots.

[0027] As shown in FIG. 7, when a nozzle allocation process is performed for printing a dot pattern (FIG. 6(a)) that does not take into account the allocation of dots to each nozzle row using multiple nozzle rows, it can be confirmed that a bias occurs in the number of ejected dots for each nozzle row. In the example of FIG. 7, the number of ejected dots is biased between a minimum of 0 dots and a maximum of 5 dots. If such a bias occurs, the life of the nozzles that frequently eject ink is shortened, resulting in a problem that the life of the print head itself is shortened. To address this problem, in this embodiment, a quantization process is performed using a dither matrix that takes into account the allocation of dots to each nozzle row, thereby generating a pattern in which the bias in the number of ejected dots for each nozzle row after nozzle allocation is suppressed.

[0028] (How to create a dither matrix) A method for generating a dither matrix used in this embodiment will be described. In the following description, the dither matrix during or after generation is designated M. The dither matrix M is a two-dimensional array with a size of Sx rows in the X direction (transport direction) and a size of Sy columns in the Y direction (recording medium width direction) (Sx, Sy are natural numbers). The size (Sx, Sy) of the dither matrix M is arbitrary, but it is desirable that the size Sx in the X direction (transport direction) be a multiple of the number of nozzle rows. Below, an example will be described in which the number of nozzle rows is 8, Sx is 64 pixels, and Sy is 64 pixels.

[0029] The Void & Cluster method is known as a method for generating a dot-dispersed dither matrix. The Void & Cluster method applies a low-pass filter to obtain a smoothed density image, and determines the arrangement of dots to be added so as to suppress local density fluctuations based on the smoothed density image, and generates a dither matrix that expresses each tone by repeating this process. In this embodiment, a similar method is used to generate a dither matrix having dot-dispersed blue noise characteristics (or green noise characteristics).

[0030] The dot pattern generated in the iterative process for generating the dither matrix M is d(x, y). The dot pattern d(x, y) is a two-dimensional array, and its size is the same as that of the dither matrix M(x, y). The value of each pixel in the dot pattern d(x, y) is 1 when a dot exists, and 0 when no dot exists. The dot pattern d(x, y) changes in the iterative process. Through the iterative process, Sx×Sy+1 dot patterns are generated, ranging from a dot pattern with 0 dots to a dot pattern with Sx×Sy dots. Therefore, when the number of dots in the dot pattern d(x, y) is g, a certain point (tone value) in the iterative process can be specified by using this g. In the following description, the number of dots g is referred to as the tone value g. In the following description, the dot pattern d(x, y) when the tone value is g is also referred to as d(g, x, y), or d(g) with x and y omitted.

[0031] Also, let n(g) be the density fluctuation map of gradation value g used to evaluate the coarseness or density of dots generated in this iterative process. In the density fluctuation map n(g), the smaller the value, the lower the smoothing density and the sparser the dots are evaluated to be, and conversely, the larger the value, the higher the smoothing density and the denser the dots are evaluated to be. In the process of S103 in FIG. 8, which will be described later, dots are added to the position in the density fluctuation map n(g) where the value is the smallest, i.e., the position where the dots are evaluated to be the sparsest. This suppresses density fluctuations in the generated dot pattern, achieving low graininess.

[0032] The density fluctuation map n(g) is a two-dimensional array of the same size as the dot pattern d(g), and the value of the array changes depending on the gradation value g, just like the dot pattern d(g). In the following description, n(g) is also written as n(g, x, y). In S102 (FIG. 8) described later, the density fluctuation map n(g) is generated by filtering the dot pattern d(g). Since it is assumed that the dither matrix is ​​applied periodically to the input image, in the filtering process in S102 described later, the density fluctuation map n(g) is generated by using the result of a cyclic convolution operation between the dot pattern d(g) and a filter coefficient. The cyclic convolution operation is an operation in which a normal convolution operation is performed between the dot pattern d(g) with a periodic boundary condition set and the filter coefficient. The details of the filter coefficient used in S102 will be described later.

[0033] Next, a dither matrix generating method according to this embodiment will be described with reference to Fig. 8. Fig. 8 is a flowchart showing the overall flow of the dither matrix generating method according to this embodiment.

[0034] In step S100, an initial density fluctuation map n(0) is generated when the gradation value g is 0. As shown in formula (1), in this embodiment, a random number (rand) of small amplitude is given to the initial density map n(0) to prevent it from forming a regular pattern.

[0035]

number

[0036] Steps S101 to S106 are repeated to add dots to the dot pattern d(g). Steps S101 and S106 are loop ends, and steps S102 to S105 are repeated to change the gradation value g from 0 to g MAX The process from S102 to S105 is a process of adding one dot to a dot pattern d(g) with a tone value of g, generating a dot pattern d(g+1) with an adjacent tone, i.e., a tone value of g+1, and adding the tone value g to the dither matrix M. In this embodiment, MAX =Sx×Sy=64×64=4096. In this way, by repeating the process of adding tone values ​​g from 1 to 4096 to the dither matrix M, dot patterns and dither matrices M for all tones are generated.

[0037] The threshold matrix generating device 5 applies a filter coefficient, which will be described later, to the dot pattern d(g) corresponding to the gradation value g to calculate the density fluctuation map n(g) (S102). The threshold matrix generating device 5 calculates the density fluctuation map n(g) at the position (x,y) where the value of the density fluctuation map n(g,x,y) is the smallest among the positions of the dot pattern d(g,x,y) that is turned off. MIN , y MIN The threshold matrix generating device 5 determines the position d(g, x) corresponding to the addition position of the dot pattern d(g) (S103). MIN , y MIN The dot pattern with the dot added becomes a dot pattern d(g+1) with a gradation value of g+1. Next, the threshold matrix generating device 5 adds a dot to the dot addition position (x MIN , y MIN The value of the dither matrix M(x, y) corresponding to the threshold value g is set to the gradation value g (S104). After that, the threshold value matrix generating device 5 increments the value of the gradation value g to g+1 (S105). The threshold value matrix generating device 5 repeats the above processes of S102 to S105 until the threshold value g is g MAXBy repeating this process until the threshold value reaches 1, a threshold matrix (dither matrix) with threshold values ​​from 1 to 4096 is generated.

[0038] When the dither matrix is ​​obtained by the above-mentioned iterative process, the threshold matrix generating device 5 adjusts the range of values ​​of the dither matrix according to the range of pixel values ​​of the input image (S107). Before the process of S107, the dither matrix M(x, y) includes a range of values ​​from 1 to 4096 (m MIN From m MAX ) is stored. When the input image is 8-bit, the range of the input image is 0 to 255 (th MIN From th MAX ), even if a dither matrix in which values ​​from 1 to 4096 are stored is used, an appropriate halftone processing result cannot be obtained. Therefore, the threshold matrix generating device 5 adjusts the range of the threshold of the dither matrix M(x, y) so as to match the expected input image (S107). For example, if the range of the threshold in the dither matrix M is adjusted to th MIN From th MAX If you want to change the threshold value of the adjusted dither matrix M, the threshold value will be the integer part of a×M(x,y)+b. MAX -th MIN )÷(m MAX -m MIN ), b=th MIN -a×m MIN It is.

[0039] (Details on generating concentration fluctuation maps through filtering) In S102, a density fluctuation map n(g) at gradation g is calculated by applying a filter f to a dot pattern d(g) of gradation g. The filter f has coefficients in a two-dimensional array of the same size as the dither matrix M, and the details of the coefficients will be described later. In this embodiment, an image filtering operation is performed on the dot pattern d(g) of gradation g using the filter f to perform a convolution operation. Specifically, the value n(g, x, y) at the coordinates (x, y) of the density fluctuation map n(g) can be calculated, for example, by the following formula (2). That is, a predetermined position (ki, kj) of the filter f is matched with the dot position d(g, x, y) to be converted in the dot pattern of the gradation value g, and an operation (convolution operation) is performed using the coefficients of the filter f. Note that, in the convolution operation, for the portion where the filter f protrudes from the dot pattern d, the dot pattern is circularly referenced, and the operation is performed by connecting the dot patterns vertically and horizontally. For example, (x+i)%Sfx represents the remainder when (x+i) is divided by Sfx. In order to obtain the effect of the initial fluctuation map n(0) set to avoid regular patterns, the initial fluctuation map is added to the calculation result of formula (2).

number

[0040] (Details of filter coefficients) Next, the filter f used in generating the density fluctuation map n(g) in S102 of this embodiment will be described. The filter f is used to extract the density fluctuation map n(g) from the dot pattern d(g). As described above, the filter f has a two-dimensional array of coefficients, and is also expressed as f(fx, fy). In this embodiment, the array size of the filter f is the same as the dither matrix M, and when the filter size in the X direction (transport direction) is Sfx and the filter size in the Y direction (recording medium width direction) is Sfy, these values ​​are 64. In S103, dots are added to the dot pattern d(g) so as to mitigate the density fluctuation in the density fluctuation map n(g), thereby reducing the density of the dots and achieving low graininess. In order to achieve this favorably, it is necessary to extract the density between the dots.

[0041] In this embodiment, the filter f for extracting the density between dots is a function related to the inverse of the distance r from the dot as shown in formula (3). This makes it easier to add dots at positions far from the positions where dots are placed, and obtains a density variation map that avoids the dots being placed in a concentrated manner. In this embodiment, 1 is added to the denominator to avoid division by zero when the distance r=0. In formula (3), dx and dy indicate positions based on the center position of the filter f, respectively, and correspond to the distances in the X and Y directions from the center position of the filter f. r(dx, dy) represents the distance r from the center of the filter f to the position (dx, dy). f(fx, fy) represents the coefficient of the filter f at the position (fx, fy), and is the reference of the filter f(ki-i, kj-j) in formula (2) described above.

[0042]

number

[0043] In addition, in formula (3), x0 and y0 represent the coordinates of the center position of the filter, and x0=Sfx÷2 and y0=Sfy÷2. In addition, % in formula (3) is a remainder operator, and abs(fx-x0)%L represents the remainder when the absolute value of fx-x0 is divided by L. L is the number of nozzle rows to which nozzles are assigned by the nozzle assignment pattern. In addition, the constant values ​​(2, 1) added in condition 1 and condition 2 in formula (3) are for the constraint of dot arrangement in this embodiment. The constant values ​​are not limited to 2 and 1, and any value larger than the surrounding function values ​​can be applied. However, the constant value used when condition 1 is satisfied is larger than the constant value used when condition 2 is satisfied. As a result, in filter f, the coefficients that pass through the position of fy=y0 and are arranged in the X direction are larger than the coefficients at other positions. In addition, among the coefficients arranged in the X direction, the coefficients arranged at a predetermined interval (the interval of the number of columns L) from the position of fy=y0 are larger than the other coefficients arranged in the X direction.

[0044] As described above, in S102, the threshold matrix generating device 5 applies the filter f created by the formula (3) to the dot pattern d(g) corresponding to the gradation value g, as in the formula (2), to calculate the density fluctuation map n(g) corresponding to the gradation value g. Then, in S103, the threshold matrix generating device 5 calculates the position (x,y) where the value of the density fluctuation map n(g,x,y) is the smallest among the positions of the dot pattern d(g,x,y) that is turned off. MIN , y MIN The threshold matrix generating device 5 determines the dot addition position d(g, x) of the dot pattern d(g). MIN , y MIN) to generate a dot pattern d(g+1) with a gradation value of g+1. By using a filter coefficient with a constant value (1) added to the position fy-y0=0 as in condition 2, the value of the density variation map becomes large at the y coordinate where the dots are arranged, making it difficult for dots to be added, and it is possible to suppress bias in the frequency of dot generation for each y position. Furthermore, as in condition 1, a filter coefficient with a larger constant value (2) added to the x coordinate position where the remainder when the y coordinate where the dots are arranged is divided by the distance L(8) from the center in the X direction is 0 is used. This makes it difficult for dots to be added consecutively at positions with an 8-pixel period in the X direction, making it possible to suppress bias in the frequency of dot generation for each nozzle row when nozzles are assigned with a nozzle assignment pattern with an 8-pixel period.

[0045] In addition, in formula (2), the initial density map n(0) is added after performing the convolution operation with the filter coefficient, but this is not limited to this. The initial density map n(0, x, y) may be added to the dot pattern (g) and then the convolution operation with the filter coefficient may be performed. In addition, the function used for the filter f is a function related to the inverse of the distance r, but this is not limited to this. For example, a low-pass filter such as a Gaussian function capable of extracting frequency components perceived as graininess may be used. For example, a two-dimensional Gaussian function (hereinafter, referred to as 2d-gaussian(x, y)) may be used as the coefficient of the filter f instead of the inverse of the distance r. When 2d-gaussian(x, y) is used, the filter coefficient may be determined by the following formula (4) instead of formula (3).

number

[0046] According to the processing flow of Figure 8 described above, when image data (dot pattern) is divided into dot patterns for each nozzle array based on a nozzle allocation pattern, a dither matrix can be provided that generates a dot arrangement that allows recording dots to be evenly allocated to multiple nozzles.

[0047] (Characteristics of generated dot patterns) FIG. 9 is a diagram for explaining the effect obtained by a dot pattern generated using a dither matrix according to this embodiment. The dither matrix is ​​generated by the process shown in FIG. 8. FIG. 9(a) shows an example of a dot pattern before nozzle allocation generated by performing a quantization process using a 64×64 size dither matrix generated by the process shown in FIG. 8. FIG. 9(b) shows a dot pattern for each nozzle row obtained as a result of performing nozzle allocation for the dot pattern of FIG. 9(a) according to nozzle allocation patterns 1 to 3 shown in FIG. 5. FIG. 10 shows the result of counting the number of dots (number of dots in the X direction) discharged by each nozzle aligned in the Y direction based on the dot pattern of FIG. 9(b). The horizontal axis of each graph indicates the nozzle position, and the vertical axis indicates the number of dots.

[0048] 10, at least when the input image data is of uniform tone, it can be seen that the difference in the number of dots included in the range of Sx (64 in this example) in the X direction at each position in the Y direction for the dot patterns printed by each nozzle array is equivalent between the nozzle arrays. Note that, although there may be a difference in the number of dots between the nozzle arrays in cases such as when the number of dots in the nozzle allocation process is not divisible by the number of nozzle arrays, as long as the bias in the number of dots is kept below a predetermined value (for example, 1 dot), it is possible to obtain the effect of sufficiently suppressing a shortened head lifespan.

[0049] (Characteristics of power spectrum) The characteristics of the power spectrum of the dot pattern obtained by the quantization processing unit 306 of this embodiment and the nozzle allocation pattern used by the nozzle allocation processing unit 308 will be described. First, the characteristics of the power spectrum of the dot pattern generated by the threshold matrix generating device 5 using the dither matrix created by the flow shown in FIG. 8 will be described. In FIG. 11, column 11a shows dot patterns for each of the gradation values ​​of 32, 64, 128, and 192, which are generated using the dither matrix created by the flow shown in FIG. 8. Column 11b shows a power spectrum (power spectrum in frequency space) in which the dot pattern shown in column 11a is subjected to a two-dimensional Fourier transform and the power (square of the absolute value) is expressed by the shade. In the power spectrum of column 11b, the center is the DC component, the u direction corresponds to the frequency component in the Y direction of the dot pattern in column 11a, and the v direction corresponds to the frequency component in the X direction of the dot pattern in column 11a, and the whiter the part, the greater the power. Note that the DC component represents the number of gradations (number of dots), and is described as 0 because it is not related to the characteristics of the dot arrangement.

[0050] From the power spectrum in frequency space of column 11b, it can be seen that it has blue noise characteristics with the power in the low frequency band suppressed, and the power of the frequency component in the X direction corresponding to the period of the number of nozzle rows L (=8) and its multiple frequencies is suppressed. Note that in this embodiment, an example has been described in which the power spectrum has blue noise characteristics, but this is not limited to this, and it is sufficient if a power spectrum having non-periodic characteristics such as green noise characteristics can be obtained.

[0051] 11(b), if the coordinate positions with the origin of the frequency space at the center (0,0) are u and v, the frequency components of the power spectrum can be expressed as F(u,v). In this case, as shown in equation (5), the power averaged in the u-axis direction for each coordinate v is defined as the u-axis averaged power spectrum uAPS(v).

[0052]

number

[0053] Here, W is the number of pixels in F(u,v) that have the same v coordinate, and is equal to the width in the y direction of the dot pattern for which the power spectrum is calculated. In the example of Fig. 11, W is 64, which is the same as the width Sy in the y direction of the dither matrix.

[0054] Column 11c of FIG. 11 shows a graph of uAPS calculated from the power spectrum shown in column 11b. As shown in column 11c, in the dot pattern of this embodiment, the power of periodic coordinates on the v axis is suppressed. m In the filter coefficients used to create the dither matrix, the frequency of the L pixel period is restricted by the dot arrangement in the X direction as shown in equation (3). m can be calculated using the following equation (6).

[0055]

number

[0056] In formula (6), H is the width in the X direction of the dot pattern for which the power spectrum is calculated, L is the number of nozzle arrays, and Z is an integer (··-3, -2, -1, 0, 1, 2, 3, 4··). In the example of FIG. 11, H is 64, which is the same as the width Sy in the Y direction of the dither matrix, and the number L of nozzle arrays is 8, so the coordinate v where the power is suppressed is m The coordinate v that satisfies equation (6) is calculated as -24, -16, -8, 0, 8, 16, 24, and 32. m , v does not satisfy v n If we express it as uAPS(v m ) <uAPS(v n ) can be confirmed.

[0057] Next, similarly to FIG. 11, the characteristics of the power spectrum in the nozzle allocation pattern shown in FIG. 5, which is applied in this embodiment, will be described. In FIG. 12, column 12a shows the L1 allocation pattern in each of the three nozzle allocation patterns shown in FIG. 5. Column 12b shows the power spectrum in the frequency space obtained by performing a two-dimensional Fourier transform on the L1 allocation pattern shown in column 12a and expressing the power (square of the absolute value) as shading. Here, the L1 allocation pattern, which is the nozzle allocation pattern of L1, will be described, but the allocation patterns of L2 to L8 are phase-shifted in the X direction with respect to the L1 pattern, and the power spectrum has the same characteristics as the L1 allocation pattern. Column 12b of FIG. 12 shows the power spectrum (power spectrum in the frequency space) obtained by performing a two-dimensional Fourier transform on the dot pattern shown in column 12a and expressing the power (square of the absolute value) as shading. In the power spectrum, the center is the DC component, the u direction corresponds to the Y-direction frequency component of the allocation pattern of column 12a, and the v direction corresponds to the X-direction frequency component of the allocation pattern of column 12a, with the whiter the area the greater the power. Note that the DC component represents the number of dots that can be allocated, and is not related to the characteristics of the allocation arrangement, so it is explained as 0. As shown in column 12b, it can be seen that the power spectrum of the frequency space of the allocation pattern corresponds to the X-direction frequency component, and has power at a frequency corresponding to the period of the number of nozzle rows L and at a multiple of that frequency.

[0058] Column 12c in Fig. 12 is the uAPS calculated from the power spectrum in column 12b. As shown in column 12c, the nozzle allocation pattern of this embodiment has power at periodic coordinates on the v axis. m is a coordinate corresponding to the frequency of L pixel periods, which are pixels that can be assigned to nozzles in the x direction, and is the same as equation (6). In the example of FIG. 12, H is 64 and the number of nozzle rows L is 8, so the coordinate v m (=H / L×Z) is calculated from equation (6) as -24, -16, -8, 0, 8, 16, 24, and 32. The coordinate v that satisfies equation (6) is v m , v does not satisfy v nIf we express it as uAPS(v m )>0, uAPS(v n )=0.

[0059] By using the dither matrix and nozzle allocation pattern having the characteristics described above using Figures 11 and 12, the dot pattern printed on the printing medium can be evenly allocated to multiple nozzle arrays based on the nozzle allocation pattern.

[0060] <Other embodiments> In the above embodiment, an example has been described in which a binary pattern in which the value of whether or not dots can be allocated (1, 0) has a predetermined characteristic is stored for each nozzle array (L1 to L8) as the nozzle allocation pattern, as shown in FIG. 5. However, the method of allocating dots to the nozzle array is not limited to this. For example, as shown in FIG. 13, a pattern in which numerical values ​​(1 to 8) corresponding to each nozzle array (L1 to L8) are arranged may be stored. In this case, the nozzle allocation processing unit 308 generates a dot pattern for each nozzle array to be printed by referring to the numerical value of the nozzle allocation pattern at the corresponding pixel position and allocating dots to each nozzle array (L1 to L8) corresponding to the numerical value. Alternatively, for example, a binary pattern as shown in FIG. 5 may be stored as a dot expansion pattern for each nozzle array. When the dot pattern after quantization processing is on, a dot pattern for each nozzle array to be printed can also be generated by expanding the value of the dot expansion pattern of the corresponding pixel as a dot pattern for each nozzle array (L1 to L8).

[0061] In the above embodiment, an example has been described in which the quantization processing unit 306 generates a dot pattern represented by two values, that is, on / off, of the dot, but the present invention is not limited to this. The above embodiment can also be applied to a dot pattern that forms two or more dots in the same pixel. For example, an input image is divided into a plurality of data, and the plurality of data are quantized using the same threshold matrix, and the quantization results are added to obtain a dot pattern that forms two or more dots. For the dot pattern of the first dot, a nozzle allocation process based on the nozzle allocation pattern is performed as in the above embodiment. Then, for the dot pattern of the second dot, a nozzle allocation process is performed according to a nozzle allocation pattern in a phase different from that of the first dot (for example, an allocation pattern shifted by L / 2 pixels in the X direction). This allows the dot pattern to be printed on the print medium to be evenly assigned to a plurality of nozzle rows.

[0062] The present invention can also be realized by a process in which a program for implementing one or more of the functions of the above-described embodiments is supplied to a system or device via a network or a storage medium, and one or more processors in a computer of the system or device read and execute the program. The present invention can also be realized by a circuit (e.g., ASIC) that implements one or more of the functions.

[0063] The disclosure of this specification includes the following image processing device and threshold matrix generating device, as well as their methods and programs. (Item 1) An image processing device that generates data for forming an image corresponding to multi-value input image data on a recording medium using a recording head in which nozzle rows, each having a plurality of nozzles arranged in a first direction, are arranged in L rows (L is 2 or more) in a second direction perpendicular to the first direction, the image processing device comprising: a conversion means for converting the input image data into a first dot pattern representing dots to be printed by the print head based on a threshold matrix in which threshold values ​​are arranged; an allocation means for allocating dots of the first dot pattern to each of the nozzle arrays so that each nozzle of the recording head is used at a period of the L arrays, and obtaining a second dot pattern representing dots recorded by each of the nozzle arrays; An image processing device characterized in that the power spectrum in frequency space of the first dot pattern has blue noise or green noise characteristics with power suppressed in the low frequency band, and power is suppressed at a frequency corresponding to the period of the L columns in the second direction and at multiples of the frequency. (Item 2) The allocating means divides the first dot pattern into the second dot patterns to be printed by each nozzle array based on an allocation pattern that indicates whether or not dots can be allocated to each of the nozzle arrays, 2. The image processing device according to item 1, wherein the allocation pattern is an arrangement of allocation patterns that allocate dots to each nozzle array at a period of the L arrays in the second direction. (Item 3) The allocation pattern has a size Tx that is a positive integer multiple of the L columns in the second direction, 3. The image processing device according to item 2, wherein a size Sx of the threshold matrix in the second direction is a positive integer multiple of the size Tx. (Item 4) The image processing device according to item 3, wherein the size of the threshold matrix and the size of the allocation pattern are equal. (Item 5) An image processing device described in any one of items 2 to 4, characterized in that the power spectrum in frequency space of the allocation pattern has power at a frequency corresponding to the period of the L columns in the second direction and at a multiple of that frequency. (Item 6) The first dot pattern obtained by the conversion means has a power spectrum in a two-dimensional frequency space with u as the coordinate corresponding to the first direction and v as the coordinate corresponding to the second direction, a size in the v direction of the first dot pattern from which the power spectrum has been calculated is H, an average of F(u,v) at the coordinate v excluding the DC component (u,v)=(0,0) of the power spectrum is defined as the u-axis direction average power spectrum uAPS(v) of the coordinate v, and a coordinate v satisfying v=(H / L)×Z (Z is an integer) is defined as v m , v does not satisfy v n When expressed as uAPS(v m ) <uAPS(v n 6. The image processing device according to any one of items 2 to 5, wherein (Item 7) The allocation pattern is a power spectrum in a two-dimensional frequency space with u as the coordinate corresponding to the first direction and v as the coordinate corresponding to the second direction, F(u,v), H as the size in the v direction of the allocation pattern in which the power spectrum is calculated, an average of F(u,v) at the coordinate v excluding the DC component (u,v)=(0,0) of the power spectrum is an average power spectrum uAPS(v) in the u-axis direction of the coordinate v, and v=(H / L)×Z (Z is an integer) is a coordinate v satisfying the above formula: m , v does not satisfy v n When expressed as uAPS(v m )>0, uAPS(v n 7. The image processing device according to any one of items 2 to 6, wherein: (Item 8) A first generating means for generating a density variation map representing the dot density of the dot pattern of the first gradation value based on a result of filtering the dot pattern of the first gradation value; a second generating means for generating a dot pattern of a second gradation value adjacent to the first gradation value by adding a dot to a dot addition position of the dot pattern of the first gradation value that is determined based on the density variation map and the dot pattern of the first gradation value; a setting means for setting the first gradation value at a position in a threshold matrix corresponding to the additional position, the filter processing is performed by aligning a predetermined position of the filter with a dot position to be converted in the dot pattern of the first gradation value and performing a calculation using coefficients of the filter, and in the filter, coefficients arranged in a first direction passing through the predetermined position are larger than coefficients at other positions, and among the coefficients arranged in the first direction, coefficients arranged at a predetermined interval from the predetermined position are larger than other coefficients arranged in the first direction. (Item 9) The threshold matrix generating device according to Item 8, wherein the predetermined position of the filter is a position that is a center of the filter in the first direction and a second direction perpendicular to the first direction. (Item 10) The threshold matrix generating device described in item 8 or 9, characterized in that the coefficients of the filter are coefficients related to the inverse of the distance from the predetermined position, and a first predetermined value is added to the coefficients arranged in the first direction, and a second predetermined value is added to the coefficients arranged in the first direction at a predetermined interval from the predetermined position. (Item 11) The threshold matrix generating device according to item 8 or 9, characterized in that the coefficients of the filter form a low-pass filter, and a first predetermined value is added to the coefficients arranged in the first direction, and a second predetermined value is added to the coefficients arranged in the first direction at a predetermined interval from the predetermined position. (Item 12) The threshold matrix generating device according to Item 11, wherein the coefficients of the low-pass filter are set based on a two-dimensional Gaussian function. (Item 13) An image processing device according to any one of items 1 to 7; the recording head in which nozzle rows each having a plurality of nozzles arranged in the first direction are arranged in L rows (L is 2 or more) in the second direction; a driving means for driving the nozzle array of the recording head in accordance with the second dot pattern. (Item 14) An image processing method for generating data for forming an image corresponding to multi-value input image data on a recording medium using a recording head in which a nozzle row having a plurality of nozzles arranged in a first direction is arranged in L rows (L is 2 or more) in a second direction perpendicular to the first direction, the method comprising: a conversion step of converting the input image data into a first dot pattern representing dots to be printed by the print head based on a threshold matrix in which threshold values ​​are arranged; an allocating step of allocating dots of the first dot pattern to each of the nozzle arrays so that each nozzle of the recording head is used at a period of the L arrays, and obtaining a second dot pattern representing dots recorded by each of the nozzle arrays; An image processing device characterized in that the power spectrum in frequency space of the first dot pattern has blue noise or green noise characteristics with power suppressed in the low frequency band, and power is suppressed at a frequency corresponding to the period of the L columns in the second direction and at multiples of the frequency. (Item 15) A first generation step of generating a density variation map representing the dot density of the dot pattern of the first gradation value based on a result of filtering the dot pattern of the first gradation value; a second generation step of generating a dot pattern of a second gradation value adjacent to the first gradation value by adding a dot to a dot addition position of the dot pattern of the first gradation value that is determined based on the density variation map and the dot pattern of the first gradation value; a setting step of setting the first gradation value at a position of a threshold matrix corresponding to the additional position, the filter processing is performed by aligning a predetermined position of the filter with a dot position to be converted in the dot pattern of the first gradation value and performing a calculation using coefficients of the filter, and in the filter, coefficients arranged in a first direction passing through the predetermined position are larger than coefficients at other positions, and among the coefficients arranged in the first direction, coefficients arranged at a predetermined interval from the predetermined position are larger than other coefficients arranged in the first direction. (Item 16) A program for causing a computer to function as each of the means of the device described in any one of Items 1 to 11.

[0064] The invention is not limited to the above-described embodiments, and various modifications and variations are possible without departing from the spirit and scope of the invention. Accordingly, the following claims are appended to apprise the public of the scope of the invention. [Explanation of symbols]

[0065] 3: image processing device, 4: printer, 5: threshold matrix generating device, 101 to 104: recording heads, 105: conveying roller, 106: recording paper

Claims

1. An image processing apparatus for generating data for forming an image corresponding to multi-valued input image data on a recording medium, using a recording head in which nozzle arrays having a plurality of nozzles arranged in a first direction are arranged in L rows (L is 2 or more) in a second direction perpendicular to the first direction, conversion means for converting the input image data into a first dot pattern representing dots to be recorded by the recording head, based on a threshold matrix in which thresholds are arranged; assignment means for assigning the dots of the first dot pattern to each of the nozzle arrays so that each nozzle of the recording head is used with the period of the L rows, and obtaining a second dot pattern representing the dots recorded by each of the nozzle arrays, wherein the power spectrum in the frequency space of the first dot pattern has blue noise or green noise characteristics in which the power in the low frequency band is suppressed, and the power is suppressed at the frequency corresponding to the period of the L rows in the second direction and its multiple frequencies. The image processing apparatus is characterized by this.

2. The assignment means divides the first dot pattern into second dot patterns to be recorded by each nozzle array, based on an assignment pattern indicating whether dot assignment to each of the nozzle arrays is possible, The image processing apparatus according to claim 1, wherein the assignment pattern has an assignment pattern for assigning dots to each nozzle array with the period of the L rows with respect to the second direction arranged.

3. The assignment pattern has a size Tx that is a positive integer multiple of the L rows in the second direction, The image processing apparatus according to claim 2, wherein the size Sx in the second direction of the threshold matrix is a positive integer multiple of the size Tx.

4. The image processing apparatus according to claim 3, wherein the size of the threshold matrix is equal to the size of the assignment pattern.

5. The image processing apparatus according to claim 2, wherein the power spectrum in the frequency space of the assignment pattern has power at the frequency corresponding to the period of the L rows in the second direction and its multiple frequencies.

6. The first dot pattern obtained by the conversion means has a power spectrum F(u, v) in a two-dimensional frequency space where the coordinate corresponding to the first direction is u and the coordinate corresponding to the second direction is v. Let H be the size of the first dot pattern in the v direction where the power spectrum is calculated. The average of F(u, v) at the coordinate v excluding the DC component (u, v) = (0, 0) of the power spectrum is defined as the u-axis direction average power spectrum uAPS(v) of the coordinate v. When the coordinate v satisfying v = (H / L)×Z (Z is an integer) is represented as vm and the coordinate v not satisfying it is represented as vn, the image processing apparatus according to claim 2, characterized in that uAPS(vm) < uAPS(vn).

7. The assignment pattern has a power spectrum F(u, v) in a two-dimensional frequency space where the coordinate corresponding to the first direction is u and the coordinate corresponding to the second direction is v. Let H be the size of the assignment pattern in the v direction where the power spectrum is calculated. The average of F(u, v) at the coordinate v excluding the DC component (u, v) = (0, 0) of the power spectrum is defined as the u-axis direction average power spectrum uAPS(v) of the coordinate v. When the coordinate v satisfying v = (H / L)×Z (Z is an integer) is represented as vm and the coordinate v not satisfying it is represented as vn, the image processing apparatus according to claim 2, characterized in that uAPS(vm) > 0 and uAPS(vn) = 0.

8. First generation means for generating a density variation map representing the dot density of the dot pattern of the first gradation value based on the result of filtering the dot pattern of the first gradation value; Second generation means for generating a dot pattern of a second gradation value adjacent to the first gradation value by adding dots at the dot addition positions determined based on the density variation map and the dot pattern of the first gradation value of the dot pattern of the first gradation value; Setting means for setting the first gradation value at the position corresponding to the addition position of the threshold matrix, and having The filter processing performs an operation using the coefficients of the filter by aligning a predetermined position of the filter with a dot position to be converted in the dot pattern of the first gradation value. In the filter, the coefficients arranged in the first direction passing through the predetermined position are larger than the coefficients at other positions, and among the coefficients arranged in the first direction, the coefficients arranged at a predetermined interval from the predetermined position are larger than the other coefficients arranged in the first direction. A threshold matrix generation device characterized by this.

9. The threshold matrix generation device according to claim 8, wherein the predetermined position of the filter is a position that is central in the first direction of the filter and in a second direction perpendicular to the first direction.

10. The coefficients of the filter are coefficients related to the reciprocal of the distance from the predetermined position, and a first predetermined value is added to the coefficients arranged in the first direction, and a second predetermined value is added to the coefficients arranged at a predetermined interval from the predetermined position among the coefficients arranged in the first direction. The threshold matrix generation device according to claim 8, characterized by this.

11. The coefficients of the filter constitute a low-pass filter, and a first predetermined value is added to the coefficients arranged in the first direction, and a second predetermined value is added to the coefficients arranged at a predetermined interval from the predetermined position among the coefficients arranged in the first direction. The threshold matrix generation device according to claim 8, characterized by this.

12. The threshold matrix generation device according to claim 11, wherein the coefficients of the low-pass filter are set based on a two-dimensional Gaussian function.

13. An image processing apparatus according to any one of claims 1 to 7, A recording head having a nozzle array in which a plurality of nozzles arranged in the first direction are arranged in L rows (L is 2 or more) in the second direction, A recording apparatus comprising: driving means for driving the nozzle array of the recording head according to the second dot pattern.

14. An image processing method for generating data for forming an image corresponding to multi-valued input image data on a recording medium using a recording head in which a nozzle array having a plurality of nozzles arranged in a first direction is arranged in L rows (L is 2 or more) in a second direction perpendicular to the first direction, A conversion step of converting the input image data into a first dot pattern representing dots recorded by the recording head based on a threshold matrix in which thresholds are arranged, An assignment step of assigning dots of the first dot pattern to each of the nozzle arrays so that each nozzle of the recording head is used at the period of the L columns, and obtaining a second dot pattern representing dots recorded by each of the nozzle arrays. The power spectrum in the frequency space of the first dot pattern has blue noise or green noise characteristics in which the power in the low frequency band is suppressed, and the power is suppressed at the frequency corresponding to the period of the L columns in the second direction and its multiple frequencies. An image processing method characterized by that.

15. A first generation step of generating a density variation map representing the density of dots of the dot pattern of the first gradation value based on the result of filtering the dot pattern of the first gradation value. A second generation step of generating a dot pattern of a second gradation value adjacent to the first gradation value by adding dots to the additional positions of the dots determined based on the density variation map and the dot pattern of the first gradation value of the dot pattern of the first gradation value. A setting step of setting the first gradation value at the position corresponding to the additional position of the threshold matrix. The filtering process performs an operation using the filter coefficients with the predetermined position of the filter aligned with the dot position to be converted in the dot pattern of the first gradation value. In the filter, the coefficients arranged in the first direction passing through the predetermined position are larger than the coefficients at other positions, and the coefficients arranged at a predetermined interval from the predetermined position among the coefficients arranged in the first direction are larger than the other coefficients arranged in the first direction. A threshold matrix generation method characterized by that.

16. A program for causing a computer to function as each means of the image processing apparatus according to any one of Claims 1 to 7.

17. A program for causing a computer to function as each means of the threshold matrix generation apparatus according to any one of Claims 8 to 12.

Images