JP2018056832A - Image processing device, image processing method, and computer program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce an influence on image quality deterioration while obtaining an effect of edge enhancement of thin lines and characters having low density.SOLUTION: An image processing device generating dot data of an output gradation value from image data of an input gradation value of an ink color, comprises a halftone processing section executing halftone processing by an error diffusion method by using a correction gradation value and a threshold value of a target pixel of the image data. The correction gradation value can be obtained by adding an error diffused from a pixel having been subjected to the halftone processing to an input gradation value of the target pixel. When diffusing an error between the correction gradation value and the output gradation value to a peripheral pixel, the halftone processing section doses not diffuse a part of the error and discards it when the input gradation value of the target pixel is at least zero. The threshold value when the input gradation value of the target pixel is included in a predetermined low gradation value range is smaller than a central value of a gradation value range which the input gradation value of the target pixel can take, and is a value equal to the input gradation value of the target pixel or more.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a halftone processing technique.

  Conventionally, an error diffusion method is known as a so-called halftone technique for converting multi-tone ink color image data into on / off data of each ink color dot. In the error diffusion method, the corrected gradation value obtained by adding the error diffused from the surrounding pixels of each pixel to the input gradation value of each pixel is compared with the threshold value, and the corrected gradation value is larger than the threshold value. In the binarization process, the dots are turned on and the dots are turned off when the correction gradation value is smaller than the threshold value. Then, an error (an error between the correction gradation value of each pixel and an output gradation value indicating ON / OFF of a dot) generated in the binarization process is diffused to pixels that have not been binarized among the peripheral pixels. This is performed while sequentially changing the target pixel. According to the error diffusion method, continuous tone control is possible while maintaining high resolution. In Patent Document 1, when an input gradation value of a certain pixel is smaller than a predetermined value, a threshold value used for binarization is set as a threshold value that is smaller than the gradation center value, and the input gradation value is large. In some cases, a technique is disclosed in which accumulation of errors caused by binarization is eliminated and a problem such as so-called delay and tailing is suppressed by setting a threshold value larger than the median gradation value as the threshold value.

Japanese Patent No. 3360391

  For example, for an image including a low density thin line on a white background such as CAD data, the average value of errors between the correction gradation value and the output gradation value described in FIG. When the error diffusion method with the threshold optimized is applied, the input gradation value is small in the thin line portion, and therefore the threshold used for binarization becomes smaller than the gradation central value. For this reason, even when the concentration of the fine line is low, the fine line is prevented from being interrupted or disappearing. If the density of the fine lines is even lower, the fine lines may still be interrupted or disappear, but in order to improve them, an overcorrected state is set to a threshold value smaller than the optimum threshold value described in FIG. A technique is also presented, which is very effective in character / line-based CAD applications. However, when an extreme overcorrection state is applied to an image in which characters / lines and photographs are mixed, edge enhancement is excessive for portions other than characters / lines. Further, if the threshold value is made too small in order to give the character portion a sufficient edge enhancement effect, image quality degradation such as tailing described in Patent Document 1 will occur. Therefore, there is a demand for a technique for reducing the influence on image quality degradation while obtaining the edge enhancement effect of a low-density thin line in a white background image.

  SUMMARY An advantage of some aspects of the invention is to solve at least a part of the problems described above, and the invention can be implemented as the following forms.

  (1) According to an embodiment of the present invention, an image processing apparatus is provided. This image processing apparatus is an image processing apparatus for generating dot data composed of output gradation values corresponding to the presence or absence of dot formation in a printing apparatus from image data composed of input gradation values of ink colors. A halftone that sequentially identifies a plurality of pixels constituting the pixel as a target pixel, and determines dot formation by an error diffusion method using a corrected gradation value obtained by correcting the input gradation value of the target pixel and a threshold value A correction tone value obtained by adding an error diffused from the pixel on which the halftone process has been performed to the target pixel to the input tone value of the target pixel. The halftone processing unit diffuses an error between the corrected gradation value and the output gradation value to a plurality of peripheral pixels that have not performed the halftone processing around the target pixel. When the input tone value of the target pixel is at least zero, a part of the error is rejected without being diffused; the threshold is determined according to the input tone value of the target pixel. The threshold value when the input gradation value of the pixel of interest is included in a predetermined low gradation value range is the gradation value range that the input gradation value of the pixel of interest can take And a value equal to or greater than the input gradation value of the target pixel.

  According to the image processing apparatus of this aspect, in the halftone process, when the error between the corrected gradation value of the target pixel and the output gradation value corresponding to the presence / absence of dot formation is diffused to the peripheral pixels, the target pixel When the input tone value of is at least zero, a part of the error is rejected without being diffused to the surrounding pixels. Since the error disappears, the accumulation error disappears at a small pixel interval compared to the configuration in which the error is diffused as it is to the surrounding pixels, and the influence of the accumulation error on the image of the surrounding pixels can be suppressed. Therefore, a negative error that occurs when the correction gradation value of a certain pixel exceeds the threshold and is determined to be “with dot formation” can be eliminated at a small pixel interval. On the other hand, the error disappearance effect also occurs in the case of a plus error. Therefore, on the contrary to the speed of disappearance of the minus error, the pixel interval until it is determined that “dot formation is present” increases due to the accumulation of the plus error. In other words, the interval at which the “dot formed” pixel occurs becomes shorter when the former effect becomes larger, and becomes longer when the latter effect becomes larger.

  However, according to the image processing apparatus of this embodiment, the threshold value when the input tone value of the target pixel is included in the low tone value range is set to be greater than the median value of the tone value range that the input tone value can take. Therefore, compared to the case where the threshold value is the median value of the gradation value range that the input gradation value can take, it is determined that “dot formation is present” with a small correction gradation value, and the absolute value is larger. Negative error occurs. For this reason, before the negative error changes from zero to positive and reaches the threshold value, the interval where the error is negative becomes longer than when the threshold value is the median value of the gradation value range that the input gradation value can take. The interval in which the error is positive is shorter than when the threshold value is the median value of the gradation value range that the input gradation value can take. For this reason, the interval in which the effect of increasing the disappearance rate of the minus error occurs becomes longer, and the interval in which the effect of reducing the accumulation rate of the plus error becomes shorter. As a result, the effect of accelerating the disappearance speed of the minus error is superior in total, and the pixel interval determined to be “with dot formation” is shortened, so that it is easy to determine “with dot formation”. This effect increases as the threshold value decreases. Even if the threshold value is simply reduced, the effect of enhancing the edge by increasing the dot generation can be obtained. However, with this type of image processing apparatus, the effect can be further increased when the background is white. For example, the edge enhancement effect can be further enhanced by using a threshold value obtained by the equation (DATA × 7 + 128) / 8 described in FIG. Although there is a slight deviation from the optimum threshold value described in FIG. 7 of Patent Document 1, image quality degradation such as tailing is suppressed to a level that does not cause a problem. Therefore, in an image including a low density fine line on a white background such as CAD data, it is possible to suppress the fine line from being interrupted or disappearing.

  As described above, according to the image processing apparatus of this embodiment, in the white background image frequently used in CAD or the like, the effect of emphasizing the low-density thin line and the character edge is obtained, and the portion other than the white background is obtained. It is possible to reduce the influence on image quality degradation caused by excessive edge enhancement.

(2) In the image processing apparatus of the above aspect, a value of an error that is diffused to the plurality of peripheral pixels without being rejected among errors between the corrected gradation value and the output gradation value is defined as a diffusion error amount. A difference between the error between the correction gradation value and the output gradation value and the diffusion error amount is set as a rejection error amount; the rejection error amount / (the correction gradation value and the output gradation value Is the rejection rate; the rejection rate when the input tone value of the pixel of interest is zero is an intermediate tone value range in which the input tone value of the pixel of interest is predetermined. It may be larger than the minimum value of the rejection rate.
According to the image processing apparatus of this aspect, since the rejection rate when the input tone value of the target pixel is zero, the input tone value of the target pixel is larger than the minimum value of the rejection rate in the intermediate tone value range, When the input gradation value is zero, the absolute value of the error diffusing to the surrounding pixels is smaller than when the input gradation value is in the intermediate gradation value range. For this reason, the amount of error (absolute value) that disappears each time error diffusion becomes larger than when the input gradation value is at the intermediate gradation value, so that the accumulation error is eliminated at a small pixel interval. Can do.

(3) In the image processing apparatus of the above aspect, the threshold value when the input gradation value of the target pixel is included in the low gradation value range is the correction gradation value and the output gradation value. It may be a value equal to or less than a threshold value set so that the average value of the errors is zero.
According to the image processing apparatus of this aspect, the average value of the error between the corrected gradation value and the output gradation value is zero when the input gradation value of the target pixel is included in the low gradation value range. Therefore, for example, a threshold optimized according to the input tone value of the target pixel such as the threshold described in FIG. 7 of Patent Document 1 can be used. Even when the image is included in a low gradation value range such as a low-density thin line or a character in a white background image, the occurrence of image quality degradation such as tailing can be suppressed.

(4) In the image processing apparatus of the above aspect, the halftone processing unit diffuses a value obtained by multiplying the error by a weight value when diffusing the error to the plurality of peripheral pixels; The tone processing unit refers to an error diffusion matrix in which the weight values are associated with the plurality of peripheral pixels, respectively, and multiplies the error by the weight value to thereby calculate the plurality of peripheral pixels. An error calculation unit that calculates an error to be diffused; in the error diffusion matrix set in advance when the input tone value of the target pixel is zero, the total value of the weight values is 1 May be a small value.
According to the image processing apparatus of this aspect, since the total value of the weight values when the input tone value of the target pixel is zero is a value smaller than 1, an error occurs when the input tone value of the target pixel is zero. It is possible to easily realize that a part of the image is rejected without being diffused to a plurality of peripheral pixels. Therefore, halftone processing can be executed in a short time, or the load required for such processing can be reduced. Further, by making the total value of the weight values when the input gradation value is zero smaller than 1, the edge enhancement effect can be enhanced. For this reason, when the total value of the weight values is made slightly smaller than 1, for example, even when the total value of the weight values is set to 0.99, the effect of edge enhancement at a level where the difference can be easily understood. Can get to.

  The present invention can be realized in various forms. For example, the present invention can be realized in the form of a printing apparatus to which the image processing apparatus is applied, a printing control apparatus that controls the printing apparatus, a computer program for realizing these apparatuses and methods, a recording medium on which such a computer program is recorded, and the like. Further, the present invention is not limited to a printing apparatus, and can be realized as an arbitrary liquid ejecting apparatus.

  A plurality of constituent elements of each aspect of the present invention described above are not indispensable, and some or all of the effects described in the present specification are to be solved to solve part or all of the above-described problems. In order to achieve the above, it is possible to appropriately change, delete, replace with another new component, and partially delete the limited contents of some of the plurality of components. In order to solve part or all of the above-described problems or to achieve part or all of the effects described in this specification, technical features included in one embodiment of the present invention described above. A part or all of the technical features included in the other aspects of the present invention described above may be combined to form an independent form of the present invention.

1 is a block diagram illustrating a configuration of an image processing apparatus as an embodiment of the present invention. It is a flowchart which shows the process sequence of a printing process. It is a flowchart which shows the detailed procedure of the halftone process (step S110) of one color component of four ink colors C, M, Y, and K. It is explanatory drawing which shows typically the relationship between the input gradation value of a focused pixel, and a threshold value. It is explanatory drawing which shows an example of a 1st error diffusion matrix. It is explanatory drawing which shows an example of a 2nd error diffusion matrix. It is explanatory drawing which shows an example of the 1st error diffusion matrix of the 1st aspect in the modification 1. It is explanatory drawing which shows an example of the 2nd error diffusion matrix of the 1st aspect in the modification 1. It is explanatory drawing which shows an example of the 1st error diffusion matrix of the 2nd aspect in the modification 1. It is explanatory drawing which shows an example of the 2nd error diffusion matrix of the 2nd aspect in the modification 1. It is explanatory drawing which shows an example of the 2nd error diffusion matrix of the 3rd aspect in the modification 1.

A. Embodiment:
A1. Configuration of image processing device:
FIG. 1 is a block diagram showing a configuration of an image processing apparatus 100 as an embodiment of the present invention. The image processing apparatus 100 performs image processing on an image printed by the printing apparatus 200, and transmits image data for printing after image processing to the printing apparatus 200. In the present embodiment, the image processing apparatus 100 is configured by a personal computer. An operating system that operates in the image processing apparatus 100 includes an application 20 and a printer driver 30. The printing apparatus 200 is configured as an ink jet printer that ejects four colors of ink, specifically, C (cyan), M (magenta), Y (yellow), and K (black) inks. Since the printing apparatus 200 has the same configuration as a general inkjet printer, a detailed description thereof is omitted.

  The printer driver 30 includes a resolution conversion unit 31, a color conversion unit 32, a halftone processing unit 33, and a print data output unit 35.

  The resolution conversion unit 31 has a function of converting the resolution of the input image into the print resolution. The input image corresponds to, for example, an image generated by the application 20 or an image received from an input interface (not shown). In the present embodiment, the input image is represented by input gradation values composed of red (R), green (G), and blue (B) color components. In the present embodiment, the gradation value range of the input image is 0 to 255.

  The color conversion unit 32 has a function of converting image data of an input image into image data composed of input gradation values of ink colors C, M, Y, and K. In the present embodiment, the input gradation value range of the image data is 0 to 255. The input gradation value range of the image data is not limited to 0 to 255, and may be an arbitrary range such as 0 to 128. The color conversion unit 32 performs the color conversion with reference to a color conversion table 36 prepared in advance.

  The halftone processing unit 33 sets the input gradation value of 256 gradations (steps) of each color of the image data to an output gradation value corresponding to the presence or absence of formation of dots of each color, in other words, dot formation present (255). A halftone process for converting to a gradation value of two gradations without dot formation (0) is performed independently for each color component of each ink color. In the present embodiment, the halftone process is executed by a so-called error diffusion method. The halftone process of the color component C will be described as an example. In the error diffusion method, a plurality of pixels constituting the image data of the color component C are specified as the target pixel in order, and the correction gradation value and the output gradation of the target pixel are specified. The presence / absence of dot formation in the target pixel is determined while diffusing an error from the value to a plurality of peripheral pixels. In the present embodiment, the “corrected gradation value” is a gradation value obtained by correcting an input gradation value, and an error diffused from a pixel that has been subjected to halftone processing to a target pixel is defined as an input gradation of the target pixel. It is a gradation value obtained by adding to the value. Further, “peripheral pixels” are pixels around the pixel of interest, and mean a plurality of pixels that have not been subjected to halftone processing.

  The halftone processing unit 33 includes an error calculation unit 34. The error calculation unit 34 calculates an error to be diffused when diffusing an error between the correction gradation value of the target pixel and the output gradation value to each of a plurality of peripheral pixels. The error to be diffused is calculated with reference to an error diffusion matrix in which respective weight values are associated with surrounding pixels. The error diffusion matrix is stored in the memory 40 in the image processing apparatus 100. In the present embodiment, as the error diffusion matrix, the case where the input tone value of the image data of the color component C of the target pixel is not zero and the case where the input tone value of the image data of the color component C of the target pixel is zero. Different error diffusion matrices are set in advance. When the input tone value of the image data of the color component C of the target pixel is not zero, the error calculation unit 34 refers to the first error diffusion matrix 37 and inputs the input tone of the image data of the color component C of the target pixel. If the value is zero, the second error diffusion matrix 38 is referred to. Although the detailed description of the first error diffusion matrix 37 and the second error diffusion matrix 38 will be described later, the second error diffusion matrix 38 is used when the target pixel is white data. The same processing as the halftone processing of the color component C described above is performed similarly for each of the remaining color components M, Y, and K.

  The print data output unit 35 is data after processing by the halftone processing unit 33, that is, data indicating whether or not each ink color C, M, Y, K is formed (hereinafter referred to as “halftone data”). Is rasterized, print data including a print control command is generated, and transmitted to the printing apparatus 200. When the halftone data is rasterized, the print data output unit 35 determines whether the dot of each pixel is formed when the carriage (not shown) in the printing apparatus 200 moves forward or double.

  The printer driver 30 corresponds to a program for realizing the functions of the processing units 31 to 35 described above. The printer driver 30 is supplied in a form recorded on a computer-readable recording medium. As such a recording medium, for example, data such as a DVD, a flash memory, an internal storage device of a computer (memory such as RAM and ROM), and an external storage device can be fixed, not temporarily, and can be read by a computer. Various media can be used.

A2. Printing process:
FIG. 2 is a flowchart illustrating a processing procedure of print processing executed by the image processing apparatus 100 and the printing apparatus 200. As shown in FIG. 2, when the user designates a print target image and issues a print instruction in the image processing apparatus 100, the print process is executed. The resolution conversion unit 31 performs a resolution conversion process on the designated print target image (step S100). Specifically, the resolution conversion unit 31 converts the resolution of the image to be printed into the resolution for printing on the print medium P.

  The color conversion unit 32 executes a color conversion process, and converts the input image data into image data composed of input gradation values of ink colors (step S105).

  The halftone processing unit 33 performs halftone processing for each color component of each ink color using the first error diffusion matrix 37 and the second error diffusion matrix 38 (step S110). Details of the halftone process will be described later.

  The print data output unit 35 executes print data output processing using the halftone data (step S115). Specifically, the print data output unit 35 performs a rasterization process on the image data after the halftone process, and outputs the processed data (print data) to the printing apparatus 200. In the printing apparatus 200, when the print data is received from the print data output unit 35, the print target image is printed on the print medium P in accordance with the control command included in the print data (step S120).

A3. Halftone processing:
FIG. 3 is a flowchart showing a detailed procedure of halftone processing (step S110) for one color component of the four ink colors C, M, Y, and K. As described above, the halftone process is started when image data of each color component of ink colors C, M, Y, and K is received from the color conversion unit 32 after the color conversion process (step S105) by the color conversion unit 32 is completed. The The halftone processing of each color component of ink colors C, M, Y, and K can be executed independently, and the halftone processing section shown in FIG. 3 is prepared for the number of ink colors and executed simultaneously in parallel for four colors. Alternatively, for example, one color may be executed in order for each line or for each screen. In the following description, the halftone process of the color component C will be described as an example.

  The halftone process is a process for generating dot data by specifying a plurality of pixels constituting the color component C image data as pixels of interest one by one in a predetermined order. Specifically, starting from the pixel at the upper left corner of the image data of the color component C, it moves one by one in the right direction, and when it reaches the right end, it moves repeatedly to the pixel at the left end one step below. In this process, the target pixel is sequentially identified and dot data is generated.

  The halftone processing unit 33 determines whether or not the processing of all the pixels of the image data of the color component C has been completed (step S205). At this time, the halftone processing unit 33 determines whether the position of the target pixel has reached the end of the image data of the color component C. When the position of the target pixel has reached the end of the image data of the color component C, it is determined that the processing for all the pixels has been completed (step S205: YES), and the halftone process for one color is ended. When it is determined that the processing for all the pixels has not been completed (step S205: NO), the next pixel is specified as the target pixel (step S210).

  After executing step S210, the halftone processing unit 33 determines whether or not the corrected gradation value of the target pixel is smaller than the threshold value (step S215). In the present embodiment, the threshold value is set in advance according to the value of the input gradation value.

  FIG. 4 is an explanatory diagram schematically showing the relationship between the input tone value of the target pixel and the threshold value. In FIG. 4, the vertical axis represents the threshold value, and the horizontal axis represents the input tone value (0 to 255) of the target pixel. In FIG. 4, the case where the threshold value is approximately the center value 128 of the gradation value range that can be taken by the input gradation value of the target pixel is indicated by a broken line. FIG. 4 corresponds to FIG. 7 of Patent Document 1 (Japanese Patent No. 3360391) described above. This threshold value is a value at which the average error between the corrected gradation value and the output gradation value becomes zero, and is a threshold value optimized according to the input gradation value. Such a threshold value can be obtained by applying the technique described in Patent Document 1. Note that “average value is zero” means a wide concept including, for example, the case where the error is in the range of plus or minus 10 in addition to the case where the average value is exactly zero. As shown in FIG. 4, the threshold value when the input gradation value of the target pixel is included in the predetermined low gradation value range is approximately the gradation value range that can be taken by the input gradation value of the target pixel. Less than median (128).

  In the present embodiment, the “low gradation value range” means, for example, a value (0 to 255) that can be taken by the input gradation value of the target pixel in an application such as a low-density thin line portion on a white background such as a CAD drawing. ) In a range corresponding to 6% to 25%. Even if the tone value is the same, if the ink density or dot size is different, the output line density varies, so the optimum tone value range is set as the “low tone value range”. Is done.

  In step S215 shown in FIG. 3, the halftone processing unit 33 determines whether or not the corrected gradation value of the target pixel is smaller than the threshold value shown in FIG. If it is determined that the correction gradation value of the target pixel is equal to or greater than the threshold (step S215: NO), the dot of the target pixel is determined to be ON (step S220). At this time, the output gradation value of the target pixel is set to 255.

  After executing step S220, the halftone processing unit 33 determines whether or not the input tone value of the target pixel is zero (step S230). When the halftone processing unit 33 determines that the input tone value of the target pixel is not zero (step S230: NO), the error between the corrected tone value of the target pixel and the output tone value is diffused to surrounding pixels. (Step S235). At this time, the error calculation unit 34 calculates the error to be diffused to the surrounding pixels as follows.

  FIG. 5 is an explanatory diagram showing an example of the first error diffusion matrix 37. The first error diffusion matrix 37 is used in error diffusion (step S235) when the input gradation value of the target pixel is not zero (step S230: NO). In FIG. 5, each square represents one pixel, and the hatched square represents a target pixel. As shown in FIG. 5, in the first error diffusion matrix 37, an error generated in the target pixel is diffused to 10 pixels around the target pixel. The weight value of the pixel on the right side of the target pixel is set to 3/16. Furthermore, the weight value of the right adjacent pixel is set to 1/16. The weight values of the pixels one step below the target pixel are set to 1/16, 2/16, 3/16, 2/16, and 1/16, respectively, in order from the left end. Furthermore, the weight values of the pixels one level below are set to 1/16, 1/16, and 1/16, respectively, in order from the left end. In the first error diffusion matrix 37, the respective weight values are set so that the total weight value is 16/16, that is, 1. Therefore, when the input gradation value of the target pixel is not zero, the error generated in the target pixel is diffused as it is to the surrounding pixels without being rejected. The error calculation unit 34 calculates an error to be diffused by multiplying an error generated in the target pixel by a weight value associated with each peripheral pixel.

  When the calculation of the error to be diffused by the error calculation unit 34 is completed, the halftone processing unit 33 adds the error to be diffused to the input tone value of the peripheral pixel to obtain a corrected tone value for the peripheral pixel, and step S235 is completed. To do. However, since errors are diffused from a plurality of peripheral pixels to each pixel, the correction gradation value is determined when error diffusion from all the peripheral pixels is completed. After step S235 is executed, the process returns to step S205 as shown in FIG.

  When it is determined in step S215 described above that the corrected gradation value of the target pixel is smaller than the threshold (step S215: YES), the halftone processing unit 33 determines that the dot of the target pixel is OFF (step S225). At this time, the output gradation value of the target pixel is set to zero.

  After execution of step S225, it is determined whether or not the input gradation value of the pixel of interest is zero in the same manner as after execution of step S220 described above (step S230). If the halftone processing unit 33 determines that the input tone value of the target pixel is zero (step S230: YES), the halftone processing unit 33 rejects part of the error between the corrected tone value of the target pixel and the output tone value. Then, the rest is diffused to surrounding pixels (step S240). Specifically, a value whose absolute value is smaller than the error between the corrected gradation value of the target pixel and the output gradation value is diffused. At this time, when the error calculation unit 34 determines that the input gradation value of the target pixel is not zero, that is, unlike the execution of step S235, the error calculation unit 34 refers to the second error diffusion matrix 38 described later to each of the peripheral pixels. Calculate the error to be diffused. Then, as in step S235, the halftone processing unit 33 adds an error to be diffused to the input gradation value of the peripheral pixel to obtain a corrected gradation value in the peripheral pixel, and step S240 is completed.

  FIG. 6 is an explanatory diagram showing an example of the second error diffusion matrix 38. In FIG. 6, as in FIG. 5, each square represents one pixel, and the hatched square represents the target pixel. The second error diffusion matrix 38 is used when the input gradation value of the target pixel is zero. The second error diffusion matrix 38 is different from the first error diffusion matrix 37 in that 2/16 is set as the weight value of the pixel on the right of the target pixel. Other configurations of the second error diffusion matrix 38 are the same as those of the first error diffusion matrix 37. In the second error diffusion matrix 38, the total value of the weight values associated with the surrounding pixels is 15/16, which is a value smaller than 1. Therefore, the total value of the errors diffused to the 10 pixels around the target pixel has a smaller absolute value than the error generated in the target pixel. For this reason, a part of the error disappears at every error diffusion.

  As shown in FIG. 3, after step S240 is executed, the process returns to step S205 in the same manner as after step S235 described above, and the above-described halftone processing is completed for all the pixels of the image data of the color component C. Steps S205 to S240 are repeatedly executed. When the halftone process for the image data of the color component C is completed, the halftone process is executed for each of the remaining color components M, Y, and K as with the color component C.

A4. Effects of halftone processing in this embodiment:
According to the halftone process described above, when the error between the correction gradation value of the target pixel and the output gradation value is diffused to a plurality of peripheral pixels, the input gradation value of the target pixel is zero. Rejects part of the error and diffuses a value that is smaller in absolute value than the error to surrounding pixels, so that part of the error is not diffused to the surrounding pixels and part of the error disappears each time the error is diffused By doing so, it is possible to eliminate the accumulated error at a small pixel interval as compared with the configuration in which the error is diffused as it is to the peripheral pixels. For this reason, it is possible to eliminate a minus error that occurs when the halftone processing result is determined to be “with dot formation (255)” at a small pixel interval. On the other hand, the error disappearance effect also occurs in the case of a plus error. Therefore, on the contrary to the speed of disappearance of the minus error, the pixel interval until it is determined that “dot formation is present” increases due to the accumulation of the plus error.

  However, when the input tone value of the target pixel is included in the low tone value range, the threshold value is smaller than the median value of the input tone value range, so the threshold value is within the input tone value range. As compared with the case of the median value of, it is determined that “dot formation is present” with a small correction gradation value, and a larger minus error of the absolute value occurs. For this reason, before the minus error changes from zero to plus and reaches the threshold value, the interval in which the error is minus is longer than when the threshold value is the median value of the input gradation value range, and the error is plus. The interval is shorter than when the threshold value is the median value of the input tone value range. For this reason, the interval in which the effect of increasing the disappearance rate of the minus error occurs becomes longer, and the interval in which the effect of reducing the accumulation rate of the plus error becomes shorter. As a result, the effect of accelerating the disappearance speed of the minus error is superior in total, and the pixel interval determined to be “with dot formation” is shortened, so that it is easy to determine “with dot formation”.

  According to the image processing apparatus of the present embodiment described above, in the halftone process, when the input gradation value of the target pixel is zero, the total weight value is calculated when calculating the error to be diffused. By using the second error diffusion matrix 38 in which a value smaller than 1 is used, a part of the error generated in the target pixel is rejected, and a value having an absolute value smaller than the error is diffused to a plurality of peripheral pixels. Therefore, a part of the error is not diffused to the surrounding pixels, and a part of the error disappears every time the error is diffused. Therefore, compared to a configuration in which a minus error is diffused as it is to the surrounding pixels, the accumulation error disappears at a small pixel interval, and it is possible to suppress the accumulation error from affecting the image of the surrounding pixels. It is possible to suppress the thin line from being interrupted or disappearing in an image including. In addition, it is difficult to determine that “dot formation is present” due to the disappearance of a part of the plus error because the threshold value is set to a value smaller than the median value of the gradation value range that the input gradation value can take. Can be suppressed.

  In addition, the threshold value is also a threshold value set so that the average value of the error between the correction gradation value and the output gradation value is zero, so that the input gradation value such as a low-density thin line or a character is used. Can be suppressed even if it is included in the low gradation value range. Therefore, in the white background image, it is possible to reduce the influence on the image quality degradation due to excessive edge enhancement on the portion other than the white background while obtaining the effect of edge enhancement of the low density fine lines and characters.

B. Variations:
B1. Modification 1:
In the above embodiment, the first error diffusion matrix 37 and the second error diffusion matrix 38 that diffuse the error generated in the target pixel to the surrounding 10 pixels are used, but the present invention is not limited to this.

  FIG. 7 is an explanatory diagram showing an example of the first error diffusion matrix 37a of the first mode in the first modification. In FIG. 7, as in FIG. 5, each square represents one pixel, and the hatched square represents a target pixel. The first error diffusion matrix 37a is different from the first error diffusion matrix 37 in that an error generated in the target pixel is diffused to four pixels around the target pixel. As shown in FIG. 7, the weight value of the pixel on the right of the target pixel is set to 7/16. The weight value of the lower left pixel of the target pixel is set to 3/16, the weight value of the pixel immediately below is set to 5/16, and the weight value of the lower right pixel is set to 1/16. In the first error diffusion matrix 37a, like the first error diffusion matrix 37, the respective weight values are set so that the total value of the weight values is 16/16, that is, 1.

  FIG. 8 is an explanatory diagram showing an example of the second error diffusion matrix 38a of the first mode in the first modification. In FIG. 8, as in FIG. 5, each square represents one pixel, and the hatched square represents the target pixel. The second error diffusion matrix 38a is different from the first error diffusion matrix 37a of the first mode in Modification 1 in that 6/16 is set as the weight value of the pixel on the right of the target pixel. Other configurations of the second error diffusion matrix 38a are the same as those of the first error diffusion matrix 37a of the first mode in the first modification. In the second error diffusion matrix 38a, the total value of the weight values associated with the surrounding pixels is 15/16, which is a value smaller than 1.

  FIG. 9 is an explanatory diagram showing an example of the first error diffusion matrix 37b of the second mode in the first modification. In FIG. 9, as in FIG. 5, each square represents one pixel, and the hatched square represents a target pixel. The first error diffusion matrix 37b is different from the first error diffusion matrix 37 in that an error generated in the target pixel is diffused to 12 pixels around the target pixel. As shown in FIG. 9, the weight value of the pixel on the right of the target pixel is set to 7/48. Furthermore, the weight value of the right adjacent pixel is set to 5/48. The weight values of the pixels one step below the target pixel are set to 3/48, 4/48, 7/48, 5/48, and 3/48, respectively, from the left end. Further, the weight values of the pixels one step below are set to 1/48, 3/48, 6/48, 3/48, and 1/48 in order from the left end. In the first error diffusion matrix 37b, the respective weight values are set so that the total weight value is 48/48, that is, 1.

  FIG. 10 is an explanatory diagram showing an example of the second error diffusion matrix 38b of the second mode in the first modification. In FIG. 10, as in FIG. 5, each square represents one pixel, and the hatched square represents the target pixel. The second error diffusion matrix 38b is different from the first error diffusion matrix 37b of the second mode in Modification 1 in that 4/48 is set as the weight value of the pixel two steps below the target pixel. Other configurations of the second error diffusion matrix 38b are the same as those of the first error diffusion matrix 37b of the second mode in the first modification. In the second error diffusion matrix 38b, the total value of the weight values associated with the surrounding pixels is 46/48, which is a value smaller than 1.

  FIG. 11 is an explanatory diagram showing an example of the second error diffusion matrix 38c of the third mode in the first modification. In FIG. 11, as in FIG. 5, each square represents one pixel, and the hatched square represents the target pixel. In the second error diffusion matrix 38c, 7/16 is set as the weight value of the pixel adjacent to the right of the target pixel, and 1/32 is set as the weight value of the lower right pixel of the target pixel. Are different from the second error diffusion matrix 38a of the first mode in the first modification. Other configurations of the second error diffusion matrix 38c are the same as those of the second error diffusion matrix 38a. In the second error diffusion matrix 38c, the total weight value associated with each of the surrounding pixels is 31/32, which is a value smaller than 1.

  A configuration in which the sum of the weight values associated with each of the surrounding pixels in the second error diffusion matrix used when the input tone value of the target pixel is zero as in each of the above-described aspects is a value smaller than 1. If so, the error generated in the target pixel may be diffused to an arbitrary number of peripheral pixels, and the weight value associated with each peripheral pixel may be an arbitrary value. Even in such a configuration, an effect similar to that of the above embodiment is produced. In addition, when an error diffusion matrix that diffuses an error to a wider range of neighboring pixels is used as in the first and second aspects of the first modification, the error generated in the target pixel is reduced more quickly. be able to.

B2. Modification 2:
In the above-described embodiment, the error calculation unit 34 refers to different error diffusion matrices depending on whether the input tone value of the target pixel is zero or the input tone value is not zero. It is not limited to. For example, even when the input tone value of the target pixel is zero, the error diffusion matrix (first error diffusion matrix 37, 37a, 37b) when the input tone value of the target pixel is not zero may be referred to. In this case, the error calculation unit 34 multiplies each value obtained by multiplying the error by the weight value associated with the error diffusion matrix by a predetermined value smaller than 1, and diffuses the error. It may be calculated as a power error. Even in this configuration, since the total value (absolute value) of the diffused errors is smaller than the error (absolute value) generated in the target pixel, the same effect as in the above embodiment can be obtained. That is, generally, when the input tone value of the target pixel is at least zero, any configuration that rejects part of the error without diffusing can be applied to the present invention.

B3. Modification 3:
In the above embodiment, the second error diffusion matrix 38 is preset only when the input tone value of the target pixel is zero, but the present invention is not limited to this. The second error diffusion matrix 38 may be set not only when the input gradation value of the target pixel is zero but also when the gradation value is larger than zero and relatively small. Thereby, the range of the white background which wants to emphasize a low concentration thin line can be expanded a little. That is, in general, when the input tone value of the target pixel is zero, if the error diffusion matrix in which the total value of the weight values is associated with a value smaller than 1 is set in advance, The same effect as the embodiment is achieved.

  The second error diffusion matrix 38 may also be set when the input tone value of the target pixel is the maximum input tone value corresponding to the data with the highest density. At this time, the threshold value corresponding to the maximum input gradation value is set larger than the median value of the input gradation value range. Thereby, for example, the reproducibility of a fine line having a slightly lower density than black with a black background can be improved. The principle of this effect can be explained by inverting the logic of white and black in the present invention. Further, the second error diffusion matrix set when the input tone value of the target pixel is the maximum input tone value corresponds to each of the surrounding pixels as long as the total weight value is smaller than 1. An error diffusion matrix in which the attached weight value and the error diffusion range are different from those of the second error diffusion matrix 38 may be used.

  As described above, even when the input gradation value of the pixel of interest is larger than zero and has a relatively small gradation value, the error generated in the pixel of interest can be reduced by using the second error diffusion matrix 38. When diffusing to surrounding pixels, part of the error can be rejected without diffusing, and the same effect as in the above embodiment can be obtained.

B4. Modification 4:
In the above embodiment, the printing apparatus 200 may have at least part of the functions of the image processing apparatus 100. Further, the configuration in which the printing apparatus 200 has all the functions of the image processing apparatus 100, for example, the configuration in which the printing apparatus 200 includes the image processing apparatus 100 may be used. Also in these structures, there exists an effect similar to embodiment.

B5. Modification 5:
In the above embodiment, the printing apparatus 200 is an ink jet printer that performs printing with four colors of ink (cyan, magenta, yellow, and black), but the present invention is not limited to this. For example, the number of inks used may be different from the above. Specifically, in addition to the above-described four colors of ink, there may be seven colors including three colors of light gray, light cyan, and light magenta. Also in these structures, there exists an effect similar to the said embodiment.

B6. Modification 6:
In the above embodiment, “when the input gradation value is zero” is determined for each color component of each ink color C, M, Y, K of the image data, but the present invention is not limited to this. For example, the input gradation values of all the color components of the ink colors C, M, Y, and K of the image data may be zero. When determining whether or not the input tone value is zero for each color component, the input tone values for all color components are not necessarily zero, so the ink color can be used even when the background is other than white. In some cases, the edge enhancement effect of the present invention may occur. However, if the input tone values of all the color components are zero, the edge enhancement effect of the present invention is effective when the background is white. Can be limited.

  Alternatively, the determination may be made at the stage of image data of an input image composed of RGB before being converted into image data composed of input gradation values of ink colors. In this case, if it is known in advance that the input gradation values of all the CMYK color components are zero when all of the three color components R, G, and B have the maximum gradation value, the input composed of RGB is performed. It can be determined that the input gradation value of the image data including the input gradation value of the ink color becomes zero at the image data stage of the image.

  In addition to this, it is not always necessary to apply the present invention to all ink colors of image data composed of input gradation values of ink colors. For example, the present invention may be applied only to the K component. Further, the present invention may be applied when the image data of the input image is monochrome data and is output only with K ink. In addition, when the printing apparatus 200 performs printing with, for example, seven color inks including three low density inks of light gray, light cyan, and light magenta, the low density thin line that is to be emphasized in the present invention is mainly low density. Therefore, the present invention may be applied only to low-density ink. Also in these structures, there exists an effect similar to the said embodiment.

B7. Modification 7:
In the above embodiment, a part of the error generated in the target pixel is not rejected when the input tone value is not zero. In other words, the error rejection rate when the input tone value is not zero is zero. However, the present invention is not limited to this. For example, if the rejection rate when the input tone value is zero is larger than the minimum rejection rate when the input tone value is not zero, the rejection rate when the input tone value is not zero. The rate is not limited to zero and may be any value. Here, the above-mentioned “rejection rate” means the following values. In other words, among the errors between the corrected gradation value and the output gradation value, the error value that is diffused to the surrounding pixels without being rejected is defined as the “diffuse error amount”, and the error between the corrected gradation value and the output gradation value And the difference between the diffusion error amount and the “rejection error amount” means a value represented by the rejection error amount / (error between the corrected gradation value and the output gradation value).

  In addition, as “when the input gradation value is not zero”, when the input gradation value is in the intermediate gradation value range or when the input gradation value is near zero (for example, the input gradation value is between 1 and intermediate) And the like up to the gradation value range). In this modification, the “intermediate gradation value” means a gradation value (64 to 192) in a range corresponding to 25% to 75% with respect to a value (0 to 255) that the input gradation value of the target pixel can take. ). Hereinafter, an example of a configuration in which the rejection rate when the input tone value is zero is larger than the minimum value of the rejection rate when the input tone value is not zero will be described.

  In the above embodiment, when the input gradation value is not zero, the error to be diffused to each peripheral pixel is the first in which the total value of the weight values shown in FIG. Although the value obtained by multiplying each weight value associated with the error diffusion matrix 37 is obtained, the value below the third decimal point is further rounded down. The process of truncating the value after the third decimal place is a process aimed at simplifying the error diffusion process within a range in which the influence on the image quality can be ignored and reducing the time required for the halftone process. By the process of rounding down the values after the third decimal place, a part of the error to be diffused, that is, the error generated in the target pixel is eventually discarded. When the error generated in the pixel of interest is a negative number, the truncation process is performed so that the absolute value becomes smaller when the value after the third decimal place is rounded down.

  In this way, by truncating the third decimal place or less, the absolute value of the rejection error amount is less than 0.01 per pixel at the maximum. For this reason, when the error is diffused to a total of 10 peripheral pixels using the first error diffusion matrix 37, the absolute value of the amount of rejection error in the entire 10 pixels is less than 0.1. Here, since the rejection rate is obtained by the amount of rejection error / (error between the corrected gradation value and the output gradation value) as described above, the rejection rate is the denominator value (corrected gradation value and output gradation value). Inversely proportional to the value). Further, the value of the denominator varies according to the correction gradation value. For this reason, the rejection rate is not constant and varies for each pixel of interest. And even if the denominator value is a very small value of 1, the rejection rate is expected to be 0.1 = 10% or less, so in the normal case, the rejection rate is expected to be about 5%, half of that. The For this reason, the minimum value of the rejection rate is a value smaller than 5% in the normal case.

  On the other hand, in the second error diffusion matrix 38 set when the input gradation value is zero, as shown in FIG. 6, the total value of the weight values is 15/16. Therefore, the rejection rate is 1/16, that is, 6.25%. Since this rejection rate is larger than the rejection rate of 5% in the normal case described above, it is larger than the minimum value of the rejection rate smaller than 5%. Thus, according to the configuration described above, the rejection rate when the input tone value is zero can be made larger than the minimum value of the rejection rate when the input tone value is not zero.

  Even in the above-described configuration, the rejection rate at a certain pixel whose input gradation value is within the intermediate gradation value range may be larger than 6.25%. However, even in this case, the minimum rejection rate including other pixels with the same input gradation value is smaller than 6.25%. The same effect as the form can be obtained.

  According to the configuration of the modified example 7 described above, even when an error is rejected when the input tone value is not zero, the rejection rate when the input tone value is zero is set to zero. By making it larger than the minimum value of the rejection rate when it is not, the same effect as the above-described embodiment is obtained.

B8. Modification 8:
In the modification example 7, the rejection rate when the input gradation value is near zero is the same as the rejection rate when the input gradation value is in the intermediate gradation value range, but the present invention is not limited to this. For example, the rejection rate when the input gradation value is near zero (for example, the range where the input gradation value is 1 to 3) may be different from the rejection rate when the input gradation value is in the intermediate gradation value range. The rejection rate when the input gradation value is zero may be the same. With such a configuration, it is possible to obtain an edge enhancement effect according to the rejection rate not only when the input gradation value is zero but also when the input gradation value is near zero.

B9. Modification 9:
In the above embodiment, different error diffusion matrices are set in advance depending on whether the input gradation value is not zero or the input gradation value is zero, but the present invention is not limited to this. For example, even when the input tone value is not zero, the same error diffusion matrix (second error diffusion matrix 38) as when the input tone value is zero may be set when the input tone value is near zero. . That is, in general, when the input gradation value is zero, if an error diffusion matrix in which a total value of weight values is associated with a value smaller than 1 is preset, the same as in the above embodiment There is an effect.

B10. Modification 10:
In the embodiment and the modification, a part of the configuration realized by hardware may be replaced by software, and conversely, a part of the configuration realized by software may be replaced by hardware. Good. In addition, when part or all of the functions of the present invention are realized by software, the software (computer program) can be provided in a form stored in a computer-readable recording medium. In the present invention, the “computer-readable recording medium” is not limited to a portable recording medium such as a flexible disk or a CD-ROM, but an internal storage device in a computer such as various RAMs and ROMs, a hard disk, etc. It also includes an external storage device fixed to the computer. That is, the “computer-readable recording medium” has a broad meaning including an arbitrary recording medium in which data can be fixed instead of temporarily.

  The present invention is not limited to the above-described embodiments and modifications, and can be realized with various configurations without departing from the spirit of the present invention. For example, the technical features in the embodiments and the modifications corresponding to the technical features in each embodiment described in the summary section of the invention are to solve some or all of the above-described problems, or In order to achieve part or all of the effects, replacement or combination can be performed as appropriate. Further, if the technical feature is not described as essential in the present specification, it can be deleted as appropriate.

    DESCRIPTION OF SYMBOLS 20 ... Application, 30 ... Printer driver, 31 ... Resolution conversion part, 32 ... Color conversion part, 33 ... Halftone processing part, 34 ... Error calculation part, 35 ... Print data output part, 36 ... Color conversion table, 37 ... 1st 1 error diffusion matrix, 37a ... first error diffusion matrix, 37b ... first error diffusion matrix, 38 ... second error diffusion matrix, 38a ... second error diffusion matrix, 38b ... second error diffusion matrix, 38c ... second error Diffusion matrix 40 ... Memory 100 ... Image processing device 200 ... Printing device P ... Print medium

Claims (6)

An image processing apparatus that generates dot data composed of output gradation values corresponding to the presence or absence of dot formation in a printing apparatus from image data composed of input gradation values of ink colors,
A plurality of pixels constituting the image data are identified as target pixels in order, and the presence / absence of dot formation is determined by an error diffusion method using a corrected gradation value obtained by correcting the input gradation value of the target pixel and a threshold value. A halftone processing unit for executing halftone processing,
The corrected gradation value is obtained by adding an error diffused from the pixel on which the halftone process has been performed to the target pixel to the input gradation value of the target pixel,
The halftone processing unit, when diffusing the error between the corrected gradation value and the output gradation value to a plurality of peripheral pixels that have not been subjected to the halftone processing around the target pixel, If the input tone value of the pixel of interest is at least zero, reject part of the error without diffusing,
The threshold value is determined according to the input gradation value of the target pixel,
The threshold value when the input gradation value of the target pixel is included in a predetermined low gradation value range is a median value of the gradation value range that the input gradation value of the target pixel can take. Is smaller than the input gradation value of the target pixel,
Image processing device. The image processing apparatus according to claim 1.
Of the error between the corrected gradation value and the output gradation value, the error value diffused to the plurality of surrounding pixels without being rejected is defined as a diffusion error amount.
The difference between the error of the corrected gradation value and the output gradation value and the diffusion error amount is a rejection error amount,
When the rejection error amount / (error between the corrected gradation value and the output gradation value) is a rejection rate,
The rejection rate when the input tone value of the pixel of interest is zero is greater than the minimum value of the rejection rate in the intermediate tone value range in which the input tone value of the pixel of interest is predetermined.
Image processing device. The image processing apparatus according to claim 1 or 2,
The threshold value when the input gradation value of the target pixel is included in the low gradation value range is such that an average value of errors between the correction gradation value and the output gradation value is zero. The value is less than or equal to the set threshold value.
Image processing device. In the image processing device according to any one of claims 1 to 3,
The halftone processing unit diffuses a value obtained by multiplying the error by a weight value when diffusing the error to the plurality of surrounding pixels.
The halftone processing unit refers to an error diffusion matrix in which the weight value is associated with each of the plurality of peripheral pixels, and multiplies the error by the weight value to thereby calculate the plurality of peripheral pixels. An error calculation unit that calculates an error to be diffused to each pixel;
In the error diffusion matrix set in advance when the input gradation value of the target pixel is zero, the total value of the weight values is a value smaller than 1.
Image processing device. An image processing method for generating dot data composed of output gradation values corresponding to the presence or absence of dot formation in a printing apparatus from image data composed of input gradation values of ink colors,
(A) A plurality of pixels constituting the image data are identified as target pixels in order, and the corrected gradation value obtained by correcting the input gradation value of the target pixel and the input gradation value of the target pixel A halftone processing step for executing halftone processing for determining the presence or absence of dot formation by an error diffusion method using a predetermined threshold value,
The threshold value when the input gradation value of the target pixel is included in a predetermined low gradation value range is a median value of the gradation value range that the input gradation value of the target pixel can take. And is set to a value equal to or greater than the input gradation value of the target pixel,
The step (a)
(A1) obtaining the corrected gradation value by adding an error diffused from the pixel on which the halftone processing has been performed to the target pixel to the input gradation value of the target pixel;
(A2) When diffusing an error between the corrected gradation value and the output gradation value to a plurality of peripheral pixels that have not been subjected to the halftone processing around the target pixel, the error of the target pixel If the input tone value is at least zero, rejecting part of the error without diffusing;
including,
Image processing method. A computer program for realizing image processing for generating dot data composed of output gradation values corresponding to the presence or absence of dot formation in a printing apparatus from image data composed of input gradation values of ink colors,
A plurality of pixels constituting the image data are identified as target pixels in order, and are determined according to a corrected gradation value obtained by correcting the input gradation value of the target pixel and the input gradation value of the target pixel. The computer realizes a halftone function that executes halftone processing that determines the presence or absence of dot formation by the error diffusion method using a threshold value,
The threshold value when the input gradation value of the target pixel is included in a predetermined low gradation value range is a median value of the gradation value range that the input gradation value of the target pixel can take. And is set to a value equal to or greater than the input gradation value of the target pixel,
The halftone function is
A function of obtaining the corrected gradation value by adding an error diffused from the pixel on which the halftone processing has been performed to the target pixel to the input gradation value of the target pixel;
When the error between the corrected gradation value and the output gradation value is diffused to a plurality of peripheral pixels that have not been subjected to the halftone processing around the target pixel, the input gradation of the target pixel If the value is at least zero, a function to reject part of the error without diffusing;
including,
program.

Images