JP6688468B2 - Dither matrix generation method, dither matrix generation device, image processing device, and dither matrix generation program - Google Patents

Description

  The present invention relates to a dither matrix generation method, a dither matrix generation device, an image processing device, and a dither matrix generation program, and more particularly to a technique for generating a dither matrix.

  Conventionally, in an image processing (binarization processing) for converting image data composed of multi-gradation (for example, 256 gradations) pixel data into image data composed of binary pixel data, an error diffusion method or The systematic dither method is widely used. The error diffusion method has a problem in that the burden of arithmetic processing is heavy, and in particular, in a highlight region, a periodic pattern or edge blunting easily occurs due to an error propagation delay. On the other hand, the systematic dither method reproduces gradation by repeating the cycle of the pixel area corresponding to the size of the dither matrix. It is assumed that such a periodic pattern may cause a deterioration in image quality due to interference with the period when the pattern itself stands out or due to mechanical characteristics of the image forming apparatus.

JP, 2014-3669, A

  However, heretofore, sufficient examination has not been made on image quality deterioration due to periodic characteristics due to use of a dither matrix.

  The present invention has been made in view of such a situation, and an object of the present invention is to provide a technique that realizes suppression of image quality deterioration due to periodic characteristics due to use of a dither matrix.

  The present invention provides a method for generating a dither matrix used for halftoning multi-tone image data. The dither matrix generation method is a dither matrix composed of the same elements as a generation target dither matrix that is the generation target dither matrix, and a plurality of adjacent dithers arranged adjacent to the generation target dither matrix. A storage element is selected from a dither matrix arranging step of arranging a matrix according to the arrangement method used in the halftone processing and a plurality of threshold values to be stored in each of a plurality of elements forming the dither matrix to be generated. Is not determined and a threshold value selection step of selecting a threshold value at which dot formation is most likely to be turned on as a threshold value of interest, and a dot is formed at a pixel corresponding to a confirmed element that is an element for which the stored threshold value has been confirmed. Remaining when the threshold to be stored is an undetermined element A matrix evaluation step of calculating a visual system evaluation value for the case where dots are formed in the element of interest selected from the prime, and based on the visual system evaluation value, the visual system evaluation value is the smallest. A storage element determining step of determining an element as a storage element of the attention threshold value, the matrix evaluation step, the whole of the generation target dither matrix and at least a part of the plurality of adjacent dither matrixes around the attention element. A step of performing a convolution operation using a bandpass filter on a range including the value of the above-mentioned range to calculate the visual system evaluation value.

  The present invention provides an image processing device. The image processing apparatus includes a halftone processing unit that performs halftone processing on multi-tone image data, and the halftone processing unit uses the dither matrix generated by the dither matrix generation method to perform the halftone processing. To execute.

  The present invention provides an apparatus for generating a dither matrix used for halftoning multi-tone image data. The dither matrix generation device is a dither matrix composed of the same elements as a generation target dither matrix that is the generation target dither matrix, and a plurality of adjacent dithers arranged adjacent to the generation target dither matrix. A storage element is selected from among a plurality of threshold values to be stored in each of a plurality of elements forming the dither matrix to be generated, and a dither matrix arrangement unit that arranges a matrix by the arrangement method used in the halftone processing. Is undetermined, and a dot is formed at a pixel corresponding to a definite element that is a definite element that is a definite element of the stored threshold, and a definite element that selects a threshold value at which dot formation is most likely to be turned on. Remaining elements that are considered to have been stored and whose threshold to be stored is an undetermined element A matrix evaluation unit that calculates a visual system evaluation value for the case where dots are formed in the element of interest selected from among the elements, and based on the visual system evaluation value, the element with the smallest visual system evaluation value is selected. A storage element determining unit that determines the storage element of the target threshold value, and the matrix evaluation unit includes the entire generation target dither matrix and at least a part of the plurality of adjacent dither matrices centering on the target element. A convolution operation using a bandpass filter is executed for the range to calculate the visual system evaluation value.

  The present invention provides a dither matrix generation program for controlling a dither matrix generation device used for halftone processing of multi-tone image data. The dither matrix generation program is a dither matrix composed of the same elements as a generation target dither matrix that is the generation target dither matrix, and a plurality of adjacent dithers arranged adjacent to the generation target dither matrix. A storage element is selected from among a plurality of threshold values to be stored in each of a plurality of elements forming a dither matrix arrangement unit that arranges a matrix in the arrangement method when used in the halftone processing, and the generation target dither matrix. An undetermined threshold selection unit that selects a threshold value at which dot formation is most likely to be turned on as a focused threshold value, and that a dot is formed in a pixel corresponding to a determined element that is an element for which the stored threshold value has been determined. The remaining requirement that is considered and is an element whose threshold to be stored is undetermined A matrix evaluation unit that calculates a visual system evaluation value for the case where dots are formed in the element of interest selected from among the elements, and the visual system evaluation value is the smallest element based on the visual system evaluation value. The dither matrix generation device is caused to function as a storage element determination unit that determines as the storage element of the target threshold value, and the matrix evaluation unit centers the target element and the entire generation target dither matrix and the plurality of adjacent dither matrices. Is calculated by performing a convolution operation using a bandpass filter on a range including at least a part of the above.

  According to the present invention, it is possible to realize the suppression of the image quality deterioration due to the periodic characteristic due to the use of the dither matrix.

3 is a block diagram showing a functional configuration of an image forming apparatus 100 and a dither matrix generation apparatus 200 according to an embodiment of the present invention. It is explanatory drawing which shows the dot pattern in the halftone process which concerns on a comparative example, and the spatial frequency characteristic of human vision. It is explanatory drawing which shows the dither matrix which concerns on a comparative example, and its usage method. 6 is a flowchart showing the content of a dither matrix generation procedure according to an embodiment. It is an explanatory view showing a situation of a dither matrix generation processing concerning one embodiment. It is a flow chart which shows the contents of convolution operation processing concerning one embodiment. It is explanatory drawing which shows a mode that a dither matrix is produced | generated using a DOG filter. It is explanatory drawing which shows an example of the characteristic of the DOG filter which concerns on one Embodiment. It is explanatory drawing which shows the characteristic of a human visual system.

  Hereinafter, modes for carrying out the present invention (hereinafter, referred to as “embodiments”) will be described with reference to the drawings.

  FIG. 1 is a block diagram showing a functional configuration of an image forming apparatus 100 and a dither matrix generation apparatus 200 according to an embodiment of the present invention. The image forming apparatus 100 includes a control unit 110, an image forming unit 120, a storage unit 140, and an image reading unit 150. The image reading unit 150 reads an image from a document and generates a multi-tone (for example, 256 tone) image data ID that is digital data. The image data ID may be given from a personal computer (not shown) or the like outside the image forming apparatus 100.

  The image forming apparatus 100 is connected to the dither matrix generation apparatus 200. The dither matrix generation device 200 includes a dither matrix arrangement unit 210, a target threshold value selection unit 220, a matrix evaluation unit 230, and a storage element determination unit 240. These functions will be described later.

  The dither matrix generation device 200 generates a dither matrix Md used when performing halftone processing by the systematic dither method in the image forming device 100. The image forming apparatus 100 performs halftone processing using the dither matrix Md to form an image on a print medium. The image formed on the print medium is used for evaluation of the dither matrix Md.

  The image forming unit 120 includes a color conversion processing unit 121, a halftone processing unit 122, an exposure unit 123, a photoconductor drum (image carrier) 126 that is an amorphous silicon photoconductor, a developing unit 124, and a charging unit 125. have. The color conversion processing unit 121 performs color conversion of the image data ID, which is RGB data, into CMYK data.

  The halftone processing unit 122 has a dither matrix Md. The halftone processing unit 122 performs halftone processing using the dither matrix Md to generate CMYK halftone data. The dither matrix Md is configured as, for example, a square matrix of 16 rows and 16 columns (or 8 rows and 8 columns to 256 rows and 256 columns), and each element of the matrix stores one of the threshold values of 0 to 255.

  The halftone processing unit 122 turns on the dot when the gradation value of the CMYK data is equal to or more than the threshold value, and turns off the dot when the gradation value of the CMYK data is less than the threshold value. As a result, the halftone processing unit 122 can generate CMYK halftones of two gradations from CMYK data of 256 gradations. The dither matrix Md can be generated by the method described later.

  The control unit 110 includes main storage means such as RAM and ROM, and control means such as MPU (Micro Processing Unit) and CPU (Central Processing Unit). Further, the control unit 110 has a controller function related to interfaces such as various I / O, USB (Universal Serial Bus), bus, and other hardware, and controls the entire image forming apparatus 100.

  Note that the dither matrix generation device 200 is also provided with a control unit, and constitutes a dither matrix arrangement unit 210, a target threshold value selection unit 220, a matrix evaluation unit 230, and a storage element determination unit 240.

  The storage unit 140 is a storage device including a hard disk drive, a flash memory, or the like, which is a non-transitory recording medium, and stores a control program and data for processing executed by the control unit 110. In the present embodiment, the storage unit 140 further stores the dither matrix Md used in the halftone process. The halftone processing unit 122 reads the dither matrix Md from the storage unit 140 and uses it.

  FIG. 2 is an explanatory diagram showing a dot pattern and a human visual spatial frequency characteristic in the halftone process according to the comparative example. As shown in FIG. 2A, the dither matrix according to the comparative example expresses gradation by increasing the number of dots Dt in the halftone dot growth mode from the highlight area to the middle of the halftone area. From the middle of the halftone area to the shadow area, the density is expressed by thickening the screen line Ln in the line growth mode.

  As shown in FIG. 2B, the human visual spatial frequency characteristic (Modulation Transfer Function: MTF) has a characteristic as a bandpass filter. Human vision has a characteristic as a bandpass filter having a center frequency of 3 to 5 cycle / degree with respect to luminance, and has a significantly low vision in a low frequency region of 1 cycle / degree or less and a high frequency region of 15 cycle / degree or more. It is known to have sensitivity.

  The dither matrix according to the comparative example has a relatively low gradation region that expresses the density in the halftone dot growth mode, and increases the number of dots Dt in the halftone dot growth mode having the blue noise characteristic. Express the key. As a result, in the highlight region, the spatial frequency of the image formed by the dots is increased, and the region in which the human visual sensitivity is low is used to realize the image formation in which the dots are less noticeable.

  On the other hand, the dither matrix according to the comparative example has a relatively high gradation area that expresses the density in the parallel line growth mode, and maintains the continuity of gradation in the shadow area. In the parallel line growth mode, the remarkably low frequency region where the human visual sensitivity is low in the line direction of the line is used, and the spatial frequency region where the human visual sensitivity is low in the direction perpendicular to the line of the line is used. ing. As a result, the dither matrix according to the comparative example achieves high image quality.

  FIG. 3 is an explanatory diagram showing a dither matrix Mc according to a comparative example and a method of using the dither matrix Mc. The dither matrix Mc is configured as a matrix of 64 elements in 8 rows and 8 columns, as shown in FIG. As shown in FIG. 3B, the dither matrix Mc is arranged side by side so as to include the size of the image to be reproduced for each CMYK color. The halftone processing unit 122 uses the plurality of dither matrices Mc arranged as shown in FIG. 3B to perform the halftone processing.

  FIG. 4 is a flowchart showing the content of the dither matrix generation procedure according to the embodiment. FIG. 5 is an explanatory diagram showing a state of the dither matrix generation processing according to the embodiment. This dither matrix generation procedure is a procedure for generating the dither matrix Md (see FIG. 1) according to the embodiment.

  In step S10, the dither matrix arranging unit 210 of the dither matrix generation device 200 determines the size of the dither matrix. The size of the dither matrix is determined according to, for example, the resolution of dots formed by the image forming apparatus 100. As the size of the dither matrix increases, a periodic pattern is less likely to occur, but there is a problem that its design becomes difficult.

  Therefore, the present invention can exert a remarkable effect in generating a dither matrix having a large size. However, in this example, a dither matrix having a size of 8 rows and 8 columns is generated for easy understanding of the description.

  In step S20, the dither matrix arranging unit 210 executes an arrangement process for a plurality of dither matrices. In this example, the dither matrix generation device 200 arranges a total of 36 dither matrices of 6 × 6. FIG. 5A and FIG. 5B show 6 dither matrices out of 36 dither matrices.

  In the example of FIGS. 5A and 5B, the six dither matrices are arranged without being displaced from each other, but they may be arranged at mutually displaced positions or at inclined positions. The 6 dither matrices are arranged by the same method as the arrangement method used in the halftone process.

  In step S30, the focused threshold value selection unit 220 of the dither matrix generation device 200 determines that the storage element is undetermined and the dot is selected from the plurality of threshold values to be stored in each of the plurality of elements forming the dither matrix. A threshold whose formation is most likely to be turned on is selected as the threshold of interest. Since all storage elements are undetermined, the focused threshold value selection unit 220 first selects the first threshold value (numerical value 1) as the minimum threshold value.

  Next, the storage element determination unit 240 determines the storage element of the first threshold value (numerical value 1). In this example, the storage element determination unit 240 determines the element (black element) arranged in the upper left corner of the preset dither matrix as the storage element of the first threshold value (numerical value 1) (FIG. 5 (a)). An element whose threshold value has been fixed is also called a fixed element. The confirmed element also includes an element confirmed based on a visual system evaluation value described later.

  In this way, the dither matrix generation device 200 generates the dither matrix M1. The halftone processing unit 122 always turns on a dot in a virtual pixel corresponding to the storage element of the first threshold value (numerical value 1) when the input gradation value is 1 or more using the dither matrix M1. It will be.

  In step S40, the target threshold value selection unit 220 selects, as the target threshold value, a threshold value in which the storage element is undetermined and dot formation is most likely to be turned on. The focused threshold value selection unit 220 selects the second threshold value (numerical value 2) as the second threshold value because the storage elements of the plurality of threshold values other than the first threshold value are undetermined. The storage element determination unit 240 determines the storage element of the second threshold value (numerical value 2).

  In this example, the dither matrix generation device 200 uses an isosceles triangle on the upper left side of the dither matrix M1 configured by a preset method, that is, a storage element (black element) of the first threshold value (numerical value 1). The element (hatching element) arranged near the center of gravity of is determined as the storage element of the second threshold value (numerical value 2) (see FIG. 5B).

  In this way, the dither matrix generation device 200 generates the dither matrix M2. When the input gradation value is 2 or more using the dither matrix M2, the halftone processing unit 122 always turns on a dot in a virtual pixel corresponding to the storage elements of the first threshold value and the second threshold value. Will be done.

  In step S50, the matrix evaluation unit 230 of the dither matrix generation device 200 executes convolution operation processing. FIG. 6 is a flowchart showing the content of the convolution operation processing according to the embodiment. The focused threshold value selection unit 220 selects the third threshold value (numerical value 3) as the third threshold value because the storage elements of the plurality of threshold values other than the first threshold value and the second threshold value are undetermined.

  In step S51, the matrix evaluation unit 230 selects a target element to be the target of the convolution operation processing. In this example, the target element is selected from 62 remaining elements excluding the storage elements of the first threshold value and the second threshold value out of the 64 elements of the dither matrix M2. The remaining element is an element whose stored threshold value is undetermined.

  In step S52, the matrix evaluation unit 230 turns dots on virtual pixels corresponding to the target element Et (see FIG. 7). The dither matrix generation device 200 turns dots on virtual pixels corresponding to all the target elements Et of the plurality of dither matrices M2. As a result, the dither matrix generation device 200 simulates an image when dots are turned on at the virtual pixel corresponding to the focused element Et, in addition to the dots corresponding to the storage elements of the first threshold value and the second threshold value. You can

  In step S53, the matrix evaluation unit 230 executes a convolution operation using a DOG (Difference of Gaussian) filter. FIG. 7 is an explanatory diagram showing how a dither matrix is generated using a DOG filter. In FIG. 7, the dither matrix Mt is a dither matrix that is a generation target (evaluation target), and is also called a generation target dither matrix. The dither matrix Ma is eight dither matrices arranged adjacent to the generation target dither matrix, and is also called an adjacent dither matrix.

  In FIG. 7, the matrix evaluation unit 230 executes the convolution operation of the DOG filter Fd on the focused element Et to calculate the visual system evaluation value. The DOG filter Fd is a square filter having a size four times the long side of the matrix size of the dither matrix M2. As a result, the matrix evaluation unit 230 can execute the convolution operation on the range including all of the eight adjacent dither matrices Ma.

  Accordingly, the matrix evaluation unit 230 can calculate the visual system evaluation value of the dither matrix, assuming a plurality of dither matrices arranged in the halftone process. The DOG filter Fd may be configured as a substantially circular DOG filter whose radius is the size of the long side of the matrix size of the dither matrix M2.

  In this way, the matrix evaluation unit 230 is formed by the dot pattern by also referring to the dots that are the target of the convolution operation, that is, the dots that are turned on by the eight adjacent dither matrices M2 around the dither matrix M2 that is the generation target. Images can be evaluated.

  Note that in FIG. 7, the element of interest Et of the dither matrix M2 that is the target of the convolution operation is hatched for the sake of clarity. However, in step S53, the dither matrix generation device 200 turns on dots in the virtual pixels corresponding to the target elements Et of all the dither matrices M2 as described above.

  FIG. 8 is an explanatory diagram illustrating an example of characteristics of the DOG filter Fd according to the embodiment. The DOG (Difference of Gaussian) Fd is generated by the difference between two Gaussian filters having different standard deviations σ. It is known that the DOG filter Fd has a characteristic as a bandpass filter that allows a specific frequency to pass depending on the setting of each standard deviation σ.

  FIG. 9 is an explanatory diagram showing the characteristics of the human visual system. The human visual system has a receptive field structure of retinal ganglion cells. In the receptive field structure, there are an ON center OFF peripheral type receptive field that recognizes a bright light spot and an OFF center ON peripheral type receptive field that recognizes a dot as a dark black point as central peripheral antagonistic receptive fields. The ON region is a region where excitement is observed with bright light (suppression with dark light). The OFF region is a region where excitement is observed in dark light (suppression in bright light).

  The OFF center ON peripheral type receptive field has an OFF region and an ON region arranged concentrically. The OFF region is arranged at the center of the receptive field, and the ON region is arranged around the OFF region. The ON region has an excitatory response to the bright part. The OFF region has an excitement response to a dark part (for example, a dot or a group of dots).

  It is known that the characteristic of the OFF center ON peripheral type receptive field can be approximated by a DOG function as a difference between the excitatory response in the ON region and the excitatory response in the OFF region. Further, the OFF center ON peripheral type receptive field has a characteristic as a bandpass filter having a center frequency of 3 to 5 cycle / degree with respect to luminance, as shown in FIG. 2B.

  The inventor of the present application newly obtained the above-mentioned findings and adjusted the standard deviation σ of the two Gaussian filters constituting the DOG filter Fd to have a peak (or center frequency) within the range of 3 to 5 cycle / degree. A bandpass filter is constructed. Thus, the inventor of the present application created a method for simulating the human visual system to quantify the periodic characteristics and other noises detected by the human visual system.

  In step S54, the matrix evaluation unit 230 executes a calculation result storage process. As a result of the convolution operation using the DOG filter Fd, the reproduction result of the image of the monotone input gradation value 2 in which only the dots of the storage elements of the first threshold value, the second threshold value and the element of interest Et are turned on is evaluated. It is output as a numerical value (also called the visual system evaluation value). The visual system evaluation value can be used as an indicator of noise detected by the human visual system.

  In this way, the matrix evaluation unit 230 can quantify and quantify the noise detected in the human visual system by using the DOG filter Fd as a model of the human visual system. The dither matrix generation device 200 executes such processing for all remaining elements (elements for which the threshold value is undetermined) (step S55).

  In step S56, the matrix evaluation unit 230 determines whether or not the minimum visual system evaluation value is one. This is because, if there are a plurality of minimum visual system evaluation values, the element for storing the threshold cannot be determined. The dither matrix generation device 200 advances the process to step S60 (see FIG. 4) when the minimum visual system evaluation value is one, and when the minimum visual system evaluation value is not one, The process proceeds to step S57 (see FIG. 4).

  In step S57, the matrix evaluation unit 230 executes the convolution operation using the Gaussian filter. The convolution operation using the Gaussian filter is performed on the plurality of elements having the smallest visual system evaluation value. Thereby, the dither matrix generation device 200 can evaluate a plurality of elements determined to be equivalent in the human visual model from another viewpoint of smoothing.

  The filter used instead of the DOG filter Fd is not limited to the Gaussian filter. For example, a DOG filter as a bandpass filter having a center frequency different from that of the DOG filter Fd may be used, and a specific filter having a characteristic different from that of the DOG filter Fd may be used. Further, the matrix evaluation unit 230 may execute a convolution operation using a specific filter for all remaining elements instead of the DOG filter Fd.

  However, the Gaussian filter has a characteristic as a low-pass filter, and has the advantage that it is possible to realize a more dispersed arrangement without selecting a storage element corresponding to the vicinity of dots already formed. Have Further, the Gaussian filter has an advantage that the arrangement can be extracted from a wide range by making the coefficient larger than that of the DOG filter Fd.

  In step S60, the storage element determination unit 240 determines the storage element of the Nth (third in this example) threshold value based on the visual system evaluation value. The dither matrix generation device 200 repeats such processing up to the last element of the dither matrix while increasing the threshold value by one (step S70) (step S80).

  In step S90, the dither matrix generation device 200 outputs the dither matrix Md. Accordingly, the image forming apparatus 100 can form a test pattern (not shown) using the dither matrix Md and verify the dither matrix Md with human eyes.

  As described above, the dither matrix generation device 200 according to the present embodiment quantitatively evaluates the periodic characteristic and other noises by using the DOG filter Fd capable of simulating the frequency characteristic of the human visual system. be able to. As a result, the dither matrix generation device 200 can realize the suppression of the image quality deterioration due to the periodic characteristic caused by the use of the dither matrix. Furthermore, the dither matrix generated by the dither matrix generation method of the present invention can reduce visual discomfort such as moire in the reproduction of a color image particularly in the combination of CMYK dot patterns.

  The inventor of the present application actually generated a dither matrix of 128 rows and 128 columns by using the method of the present invention. From this, the inventor of the present application has actually confirmed that the present invention has achieved the object.

  The present invention can be implemented not only in the above embodiment but also in the following modified examples.

  Modification 1 In the above embodiment, the 256 gradations are binarized using the dither matrix, but the invention is not limited to binarization. Since the dither matrix can be evaluated by using the DOG filter Fd which is a model of the human visual system, the present invention can be applied to the case where the dither matrix is converted into three or more values.

  Modified Example 2: In the above embodiment, the DOG filter is used as a filter capable of simulating the frequency characteristics of the human visual system, but the present invention is not limited to the DOG filter. Specifically, instead of the DOG filter, for example, a LOG (Laplacian Of Gaussian Filter) filter may be used.

  The LOG filter applies the Laplacian filter to the image smoothed by the Gaussian filter, and has a characteristic similar to that of the DOG filter. However, the DOG filter has an advantage that the calculation load is smaller than that of the LOG filter.

  Modification 3: In the above-described embodiment, the DOG filter Fd is a square filter having a size four times the long side of the matrix size of the dither matrix M2 with respect to the range including the entire plurality of adjacent dither matrices. Performing a convolution operation.

  However, it is not always necessary to perform the convolution operation on the range including the entire plurality of adjacent dither matrices, and the convolution is performed on the range including the entire target dither matrix and at least a part of the plurality of adjacent dither matrices. It suffices to execute the calculation. However, if the convolution operation is executed on the range including all of the plurality of adjacent dither matrices, it is possible to further suppress the image quality deterioration due to the periodic characteristic due to the use of the dither matrix. The range in which the convolution operation is executed may be expanded to the range including the dither matrix adjacent to the outside of the adjacent dither matrix.

  Modification 4: In the above embodiment, the present invention is applied to an image forming apparatus or the like that uses a dither matrix, but can also be applied to an image processing apparatus that uses a dither matrix. The image processing apparatus and the image forming apparatus using the dither matrix according to the present invention have an advantage that they can be processed at high speed without requiring a feedback configuration for diffusing an error unlike error diffusion. .

100 image forming apparatus 110 control unit 120 image forming unit 121 color conversion processing unit 123 exposure unit 124 developing unit 125 charging unit 126 photoconductor drum (image carrier)
140 storage unit 150 image reading unit

Claims (8)

A method for generating a dither matrix used for halftone processing of multi-tone image data, comprising:
A dither matrix composed of the same elements as the generation target dither matrix that is the generation target dither matrix, and a plurality of adjacent dither matrices arranged adjacent to the generation target dither matrix are subjected to the halftone processing. A dither matrix placement step that places the placement method when used,
From among a plurality of threshold values to be stored in each of the plurality of elements forming the generation target dither matrix, a threshold value for which a storage element is undetermined and dot formation is most likely to be turned on is selected as a target threshold value. A threshold selection step,
It is considered that a dot is formed in the pixel corresponding to the deterministic element that is the element whose threshold is stored, and the threshold to be stored is the element of interest selected from the remaining elements that are undetermined elements. A matrix evaluation step of calculating a visual system evaluation value when assuming that dots are formed,
Based on the visual system evaluation value, the storage element determining step of determining the element having the smallest visual system evaluation value as the storage element of the target threshold value,
Equipped with
In the matrix evaluation step, a convolution operation using a bandpass filter is performed on a range including the entire generation target dither matrix and at least a part of the plurality of adjacent dither matrices centered on the element of interest to perform the convolution operation. The visual system evaluation value is calculated, and it is determined whether or not there is one element with the smallest visual system evaluation value. When there are a plurality of elements with the smallest visual system evaluation value, the bandpass filter A dither matrix generation method including the step of calculating the visual system evaluation value using a Gaussian filter having a size wider than that of the above . The dither matrix generation method according to claim 1, wherein
The dither matrix generation method, wherein the band pass filter includes a DOG filter. The dither matrix generation method according to claim 1 or 2 , wherein
The dither matrix generation method, wherein the matrix evaluation step includes a step of calculating the visual system evaluation value by performing the convolution operation on a range including all of the plurality of adjacent dither matrices. The dither matrix generation method according to any one of claims 1 to 3 ,
The storage element determining step has a step of determining, as a preset element, an element for storing a threshold value in which dot formation is most likely to be turned on and a threshold value in which dot formation is most likely to be turned on from among the plurality of threshold values. Then
The said matrix evaluation process is a dither matrix generation method which has the process of calculating the said visual system evaluation value with respect to the said remaining element. The dither matrix generation method according to any one of claims 1 to 4 , wherein
The dither matrix generating method, wherein the bandpass filter has a peak in a range of 3 to 5 cycle / degree. An image processing device,
Equipped with a halftone processing unit that performs halftone processing on multi-tone image data,
The image processing apparatus, wherein the halftone processing unit executes the halftone processing using a dither matrix generated by the dither matrix generation method according to claim 1 . A device for generating a dither matrix used for halftone processing of multi-tone image data,
A dither matrix composed of the same elements as the generation target dither matrix that is the generation target dither matrix, and a plurality of adjacent dither matrices arranged adjacent to the generation target dither matrix are subjected to the halftone processing. A dither matrix placement part that is placed according to the placement method used,
From among a plurality of threshold values to be stored in each of the plurality of elements forming the generation target dither matrix, a threshold value for which a storage element is undetermined and dot formation is most likely to be turned on is selected as a target threshold value. A threshold selection section,
It is considered that a dot is formed in the pixel corresponding to the deterministic element that is the element whose threshold is stored, and the threshold to be stored is the element of interest selected from the remaining elements that are undetermined elements. A matrix evaluation unit that calculates a visual system evaluation value for the case where dots are formed,
Based on the visual system evaluation value, a storage element determination unit that determines the element having the smallest visual system evaluation value as the storage element of the target threshold value,
Equipped with
The matrix evaluation unit performs a convolution operation using a bandpass filter on a range that includes the entire generation target dither matrix and at least a portion of the plurality of adjacent dither matrices centering on the element of interest. The visual system evaluation value is calculated, and it is determined whether or not there is one element with the smallest visual system evaluation value. When there are a plurality of elements with the smallest visual system evaluation value, the bandpass filter A dither matrix generation device for calculating the visual system evaluation value by using a Gaussian filter having a wider range of sizes . A dither matrix generation program for controlling a dither matrix generation device used for halftone processing of multi-tone image data,
A dither matrix composed of the same elements as the generation target dither matrix that is the generation target dither matrix, and a plurality of adjacent dither matrices arranged adjacent to the generation target dither matrix are subjected to the halftone processing. The dither matrix placement part, which is placed according to the placement method used,
From among a plurality of threshold values to be stored in each of the plurality of elements forming the generation target dither matrix, a threshold value for which a storage element is undetermined and dot formation is most likely to be turned on is selected as a target threshold value. Threshold selection section,
It is considered that a dot is formed in the pixel corresponding to the deterministic element that is the element whose threshold is stored, and the threshold to be stored is the element of interest selected from the remaining elements that are undetermined elements. A matrix evaluation unit that calculates a visual system evaluation value when dots are assumed to be formed, and based on the visual system evaluation value, an element having the smallest visual system evaluation value is determined as a storage element of the attention threshold value. Causing the dither matrix generation device to function as a storage element determination unit,
The matrix evaluation unit performs a convolution operation using a bandpass filter on a range that includes the entire generation target dither matrix and at least a portion of the plurality of adjacent dither matrices centering on the element of interest. The visual system evaluation value is calculated, and it is determined whether or not there is one element with the smallest visual system evaluation value. When there are a plurality of elements with the smallest visual system evaluation value, the bandpass filter A dither matrix generation program for calculating the visual system evaluation value by using a Gaussian filter having a wider range of sizes .