JP2021122952A - Image processing device, dot image generation apparatus, printing system and image processing method - Google Patents

Abstract

To suppress unevenness in a print image.SOLUTION: An image processing device 5 comprises: a target region identification unit 51; and an image conversion unit 52. In an input image for an ink jet printer, a color at each position is expressed with the density of plural color components. The target region identification unit 51 identifies from the input image the target region where the density of one color component is 0% and unevenness is conspicuous in the print image when performing printing by the ink jet printer that uses ink in the plural colors corresponding to the plural color components. The image conversion unit 52 generates a converted image obtained by converting the color of the target region of the input image into a pseudo color by changing the density of the plural color components while making the density of the one color component larger than 0% in the target region of the input image. Consequently, unevenness in the target region in the print image can be suppressed.SELECTED DRAWING: Figure 2

Description

The present invention relates to a technique for processing an input image for an inkjet printer.

Conventionally, an inkjet printer that prints on a predetermined base material by ejecting minute droplets of ink from a plurality of ejection ports provided in a print head has become widespread. In addition, a print head in which a plurality of ejection ports are arranged over the entire width of the base material is provided, and the base material passes under the print head once to perform high-speed printing on the base material. The type inkjet printer is also used in industrial printing.

Patent Document 1 discloses an image processing device that converts a multi-valued color luminance signal into a binary or multi-valued color density control signal. In the device, when converting the input value of the color brightness signal to the density output, if the input value belongs to the highlight area, the table for density output corresponding to the first color and the second color are displayed. The interpolation calculation is performed using a table consisting of the sum of the corresponding density output tables. If the input value does not belong to the highlight area, the interpolation calculation is performed using the density output table corresponding to the first color.

Japanese Unexamined Patent Publication No. 2004-128975

By the way, in an inkjet printer, ink mist adheres to the vicinity of an ejection port (for example, an ejection port provided on a nozzle plate), and the ink is ejected in an inappropriate direction, or the ink is not ejected. Flyability may decrease. In this case, streaky unevenness occurs in the printed image. Such unevenness is particularly likely to become apparent when a flat net having a large area, a gradation, or the like is printed by using an ink having a low brightness in a single color. When comparing the four colors of C (cyan), M (magenta), Y (yellow), and K (black), the brightness of K is typically the lowest, the brightness of Y is the highest, and C and The brightness of M is almost the same.

The present invention has been made in view of the above problems, and an object of the present invention is to suppress unevenness in a printed image.

The invention according to claim 1 is an image processing apparatus that processes an input image for an inkjet printer, and the color at each position in the input image is expressed by the densities of a plurality of color components. The target area in which the density of the color component is 0% and unevenness is easily noticeable in the printed image when printed by an inkjet printer using inks of a plurality of colors corresponding to the plurality of color components is described above. The input is performed by changing the density of the plurality of color components while increasing the density of the one color component to more than 0% in the target area specifying portion specified from the input image and the target area of the input image. It includes an image conversion unit that generates a converted image obtained by converting the color of the target area of the image into a pseudo color.

The invention according to claim 2 is the image processing apparatus according to claim 1, wherein the target area specified by the target area specifying unit has a substantially constant density or a stepwise change in density. Area to do.

The invention according to claim 3 is the image processing apparatus according to claim 1 or 2, wherein the target area specified by the target area specifying unit is a region in which a single color component is used.

The invention according to claim 4 is the image processing apparatus according to any one of claims 1 to 3, and the area attribute data indicating the attributes of each area in the input image is stored in the target area identification unit. Upon input, the target area identification unit identifies the target area based on the area attribute data.

The invention according to claim 5 is the image processing apparatus according to any one of claims 1 to 3, wherein at least a part of the target area specifying portion is constructed by learning using a teacher data group. , The teacher data group includes an image to which one label indicating an attribute in which unevenness is easily noticeable is given in a printed image, and an image to which another label showing an attribute in which unevenness is less noticeable is given.

The invention according to claim 6 is the image processing apparatus according to any one of claims 1 to 5, wherein the color component having the maximum density in the target region of the input image is the color component of interest. The density of the color component of interest is greater than 0% in the target region of the converted image.

The invention according to claim 7 is the image processing apparatus according to any one of claims 1 to 5, wherein the color component having the maximum density in the target region of the input image is the color component of interest. The density of the color component of interest is 0% in the target region of the converted image.

The invention according to claim 8 is the image processing apparatus according to any one of claims 1 to 7, wherein the image conversion unit determines the pseudo color based on the type of an image in the target area. do.

The invention according to claim 9 is a halftone dot image generation device for generating a halftone dot image used in an inkjet printer, the image processing device according to any one of claims 1 to 8, and the image processing. It is provided with a shading portion that generates a halftone dot image of each color component by performing a shading process on the image of each color component of the converted image generated by the apparatus.

The invention according to claim 10 is a printing system, which is based on the halftone dot image generator according to claim 9 and the halftone dot image of each color component generated by the halftone dot image generator. It is equipped with an inkjet printer that prints on the material by the inkjet method.

The invention according to claim 11 is an image processing method for processing an input image for an inkjet printer, wherein a) the color at each position in the input image is expressed by the density of a plurality of color components. A target area in which the density of one color component is 0% and unevenness is easily noticeable in a printed image when printed by an inkjet printer using inks of a plurality of colors corresponding to the plurality of color components. , And b) in the target region of the input image, by changing the density of the plurality of color components while increasing the density of the one color component to more than 0%. The present invention includes a step of generating a converted image in which the color of the target area of the input image is converted into a pseudo color.

The invention according to claim 12 is the image processing method according to claim 11, and the density of the target region specified in the step a) is approximately constant, or the density changes stepwise. Area to do.

The invention according to claim 13 is the image processing method according to claim 11 or 12, and the target region specified in the step a) is a region in which a single color component is used.

The invention according to claim 14 is the image processing method according to any one of claims 11 to 13, wherein region attribute data indicating the attributes of each region in the input image is prepared, and the step b). In, the target area is specified based on the area attribute data.

The invention according to claim 15 is the image processing method according to any one of claims 11 to 13. In the step a), the target area is specified by the target area specifying unit, and the target area is specified. At least a part of the specific part is constructed by learning using the teacher data group, and the teacher data group includes an image with one label indicating an attribute in which unevenness is easily noticeable in the printed image and an attribute in which unevenness is less noticeable. Includes images with other labels indicating.

The invention according to claim 16 is the image processing method according to any one of claims 11 to 15, wherein the color component having the maximum density in the target region of the input image is the color component of interest. The density of the color component of interest is greater than 0% in the target region of the converted image.

The invention according to claim 17 is the image processing method according to any one of claims 11 to 15, wherein the color component having the maximum density in the target region of the input image is the color component of interest. The density of the color component of interest is 0% in the target region of the converted image.

The invention according to claim 18 is the image processing method according to any one of claims 11 to 17, wherein in the step b), the pseudo color is determined based on the type of the image in the target region. Will be done.

According to the present invention, unevenness in a target region can be suppressed in a printed image.

It is a figure which shows the structure of a printing system. It is a block diagram which shows the functional structure realized by a computer. It is a figure which shows the flow of the process of printing an image. It is a figure which shows the input image. It is a block diagram which shows another example of an image processing apparatus.

FIG. 1 is a diagram showing a configuration of a printing system 1 according to an embodiment of the present invention. In FIG. 1, arrows indicate the X, Y, and Z directions that are orthogonal to each other. Typically, the Z direction is the vertical direction, but the Z direction may be inclined with respect to the vertical direction. The printing system 1 includes a computer 11 and an inkjet printer 12. The computer 11 has a CPU that performs various arithmetic processes, a memory that stores various information, and the like. The functions realized by the computer 11 will be described later.

The inkjet printer 12 is an image recording device that superimposes and records images of a plurality of color components on a base material 9. In the example of FIG. 1, the base material 9 is a continuous sheet such as paper or film. The inkjet printer 12 includes a discharge unit 2, a transfer mechanism 3, and a control unit 4. The control unit 4 is connected to the discharge unit 2 and the transfer mechanism 3 and controls both of them. The control unit 4 is, for example, a computer having a CPU or the like. The computer 11 described above is also connected to the control unit 4.

The transport mechanism 3 includes a plurality of rollers 31, a supply section 32, and a take-up section 33. The supply unit 32 is arranged on the most (−Y) side and holds the roll-shaped base material 9 (supply roll). The take-up portion 33 is arranged on the most (+ Y) side and holds the roll-shaped base material 9 (wind-up roll). The plurality of rollers 31 are arranged between the supply unit 32 and the take-up unit 33. Each roller 31 has a central axis parallel to the X direction. The base material 9 is hung on a plurality of rollers 31 between the supply unit 32 and the take-up unit 33. In the printing system 1, a part of the rollers 31 are connected to a motor (not shown), and by rotating at a constant rotation speed around the central axis, the transfer path from the supply unit 32 to the take-up unit 33 is established. The base material 9 moves along the base material 9 at a constant speed.

The discharge portion 2 is provided on the (+ Z) side of the base material 9 in the transport path. The discharge unit 2 includes a plurality of heads 21. The plurality of heads 21 are arranged along the longitudinal direction of the base material 9. A plurality of colors of ink different from each other are supplied to the plurality of heads 21. In each head 21, a plurality of discharge ports are arranged in the X direction. The width direction of the base material 9 in the transport path is parallel to the X direction, and the plurality of discharge ports are arranged along the width direction of the base material 9. In a preferred printing system 1, the plurality of ejection ports are provided over the entire width of the base material 9. The plurality of ejection ports of each head 21 eject minute droplets of ink of one color toward the base material 9.

In the example of FIG. 1, four heads 21 are arranged in the Y direction in the discharge unit 2. The most (−Y) side head 21 ejects cyan (C) ink. Hereinafter, the head 21 is referred to as "C head 21". The same applies to the head 21 that ejects inks of other colors. The head 21 adjacent to the (+ Y) side of the C head 21 ejects magenta (M) ink, and the head 21 adjacent to the (+ Y) side of the M head 21 ejects yellow (Y) ink. do. The head 21 on the most (+ Y) side ejects black (K) ink. The arrangement order of these heads 21 may be changed as appropriate. In the following description, it is assumed that the printing system 1 has heads 21 of C, M, Y, and K. However, in the printing system 1, heads 21 of other colors may be used in accordance with the input image described later. good.

FIG. 2 is a block diagram showing a functional configuration realized by the computer 11. In the printing system 1, the functions of the image processing device 5 and the shading portion 6 in FIG. 2 are realized by the computer 11 executing a predetermined program. A part or all of the image processing device 5 and the shaded portion 6 may be realized by a dedicated electric circuit. In FIG. 2, the control unit 4 of the inkjet printer 12 is also shown.

The image processing device 5 performs predetermined processing on an input image input from the outside. The image processing device 5 includes a target area specifying unit 51 and an image conversion unit 52. The target area specifying unit 51 identifies the target area described later from the input image. The image conversion unit 52 converts the color of the target area in the input image to generate a converted image. The shading unit 6 generates a halftone dot image used in the inkjet printer 12 from the converted image.

Next, the process of printing an image by the printing system 1 will be described with reference to FIG. When the printing system 1 prints an image on the base material 9, first, the input image data 81 for the inkjet printer 12 is prepared (step S11). The input image data 81 is image data indicating a multicolor input image. In the input image, the color of each position (pixel) is represented by the densities of a plurality of color components. For example, the densities of each color component are represented by 8 bits (0 to 255). In the following description, it is assumed that each position of the input image is represented by the color components of C, M, Y, and K, but it may be represented by other combinations of two or more color components.

FIG. 4 is a diagram showing an input image 7 indicated by the input image data 81. A plurality of image areas 71 to 75 are set in the input image 7. Further, the input image data 81 is accompanied by the area attribute data 82. The area attribute data 82 indicates the position and size of each image area 71 to 75 (that is, the number of pixels in the vertical and horizontal directions), and the attributes of the image areas 71 to 75. The area attribute data 82 associates each image area 71-75 with an attribute indicating the type or property of the image.

The image area 71 in the input image 7 of FIG. 4 is associated with an attribute indicating a natural image such as a natural landscape or an artificial landscape. An attribute indicating a logo (logotype), which is a character having a relatively large size, is associated with the image area 72. An attribute indicating a binary character string having a relatively small size is associated with the image area 73. An attribute indicating a barcode is associated with the image area 74. An attribute indicating the background is associated with the image area 75. The image area 75 is an area excluding the image areas 71 to 74. The area attribute data 82 is generated, for example, by setting the positions and sizes of the image areas 71 to 75 and determining the attributes of the image areas 71 to 75 when the operator generates the input image 7. NS. In FIG. 2, the combination of the input image data 81 and the area attribute data 82 is shown as the input image data set 8.

In the target area specifying unit 51, the target area is specified from the input image 7 (step S12). Here, in the inkjet printer 12, printing is performed on the base material 9 based on the input image 7, and in the printed image, regions corresponding to the respective image regions 71 to 75 of the input image 7 are formed. The target area is an area in which unevenness is easily noticeable in (corresponding area) of the printed image when the input image 7 is printed by the inkjet printer 12.

For example, in the input image 7, the region where the density is constant is represented as a flat net region in the halftone dot image described later. Further, in the inkjet printer 12, ink is ejected from the same ejection port at each position of the base material 9 in the width direction. Therefore, in printing in a region where the density is constant, when the density in the region is high and the ink ejection frequency is relatively high, and the ink flying property of some ejection ports is low, the printed image is said to be relevant. Streaky unevenness becomes conspicuous (easily manifested) in the region. In the target area specifying unit 51, an image area including such an area is specified as a target area. On the other hand, even if the density is constant, streaky unevenness is less noticeable in areas such as barcodes, small-sized characters, and thin lines, and an image area including only the area is not a target area.

Further, in the region where inks of a plurality of colors are used in the printed image, even if the ink flying property of some of the ejection ports is deteriorated, the influence thereof is alleviated. In other words, the streak-like unevenness in the printed image becomes more remarkable in the region where the ink of a single color is used. Therefore, in the target area specifying unit 51, it is preferable that the area in which a single color component is used in the input image 7 is specified as the target area. In this processing example, an image region in which the density of K is 100% (value is 255) and the density of each of C, M, and Y is 0% (value is 0) is specified as the target region. It shall be. As will be described later, the target area is not limited to such an area.

In the target area specifying unit 51, each image area 71 to 75 of the input image 7 is specified based on the area attribute data 82 input together with the input image data 81, and the attributes of the image areas 71 to 75 are acquired. NS. Subsequently, among the plurality of image areas 71 to 75 of the input image 7, an image area having a predetermined attribute (hereinafter, referred to as “designated attribute”) is extracted as a candidate for the target area. Here, the logo and the background are designated attributes, and the image areas 72 and 75 are candidates for the target area. The image area of the designated attribute includes an area that spreads over a wide area at a constant density. In addition, other attributes such as illustrations may be included in the designated attributes.

In the target area identification unit 51, the densities of C, M, Y, and K in the image areas 72 and 75 that are candidates for the target area are confirmed, the density of K is 100%, and other colors (C, M). , Y, respectively) is 0%, the image area is specified as a target area. At this time, the blank area where all the concentrations of C, M, Y, and K are 0% is ignored. In the image area 72 in the example of FIG. 4, since the entire area excluding the blank area is C0%, M0%, Y0%, and K100%, the image area 72 is specified as the target area. Further, since the entire image area 75 is C0%, M0%, Y0%, and K100%, the image area 75 is also specified as the target area. The density of K in the image region specified as the target region does not necessarily have to be 100%, and may be a predetermined value or more (for example, 60% or more, preferably 80% or more). In the target area specifying unit 51, the image areas having C0%, M0%, Y0%, and K100% are specified first, and the image area having the designated attribute among these image areas may be specified as the target area.

When the target area is specified from the input image 7, the image conversion unit 52 generates a converted image in which the color of the target area of the input image 7 is converted into a pseudo color (step S13). In this processing example, K, which is the color of the target area, is represented by a mixture of C, M, and Y. At this time, the preferred image conversion unit 52 determines the pseudo color based on the type (attribute) of the image in the target area. For example, in the image area 75 showing the background, the entire color is converted (color separation) into C35%, M35%, Y35%, and K0%. Further, in the image area 72 showing the logo, the entire color excluding the blank area is converted into C30%, M30%, Y30%, and K10%. As described above, in the image conversion unit 52, a plurality of color components while increasing the density of the color components (here, C, M, Y) having a density of 0% in the target region of the input image 7 to more than 0%. The converted image is generated by changing the density of. The converted image is input to the shaded area 6.

In the shaded portion 6, an image of each color component is generated from the converted image (that is, the image is divided into C, M, Y, and K). Then, by performing a shading process on the image of each color component, a halftone dot image of the color component is generated (step S14). In one example of shading, a threshold matrix is prepared for each color component. The threshold matrix is a two-dimensional array of thresholds. When the image of each color component and the threshold matrix are superimposed and the value of each pixel of the image of the color component is larger than the corresponding threshold of the threshold matrix, dots are placed on the pixels in the halftone dot image. In other cases, no dots are placed. In this way, by comparing the image of each color component with the threshold matrix, a halftone dot image represented by an AM (Amplitude Modulated) screen or an FM (Frequency Modulated) screen is generated. In the shading portion 6, a shading process that does not use the threshold matrix (for example, a shading process by an error diffusion method or the like) may be performed. In the printing system 1, the halftone dot image generation device is realized by the image processing device 5 and the shading unit 6.

When halftone dot images of a plurality of color components are generated, the inkjet printer 12 prints on the base material 9 by an inkjet method based on the halftone dot images of each color component (step S15). Here, in the inkjet printer 12, inks of a plurality of colors corresponding to the plurality of color components of the input image 7 (inks of C, M, Y, and K in this processing example) are used. The halftone dot images of C, M, Y, and K are images printed on the base material 9 by the heads 21 of C, M, Y, and K, respectively, and a plurality of pixels in the halftone dot image of each color component are base materials. It is considered that they are arranged and set on 9. In the inkjet printer 12, the plurality of ejection positions on the substrate 9 corresponding to the plurality of ejection ports of the head 21 of each color component are moved relative to the substrate 9, and the plurality of ejection positions are moved according to the halftone dot image. Ink ejection from the ejection port is controlled. As a result, the printed image showing the input image 7 is printed on the base material 9, and the process in the printing system 1 is completed.

Here, a printing system of a comparative example in which the image processing device 5 is omitted is assumed. In the printing system of the comparative example, since the input image 7 is input to the shaded area 6 as it is, both the image area 72 showing the logo and the image area 75 showing the background are C0%, M0%, and so on. The shading process is performed in the state of Y0% and K100%. As a result, in the image printed by the inkjet printer 12, the logo of the image area 72 is formed only with the ink of K, and the background of the image area 75 is also formed only with the ink of K. Therefore, in the K head 21 of the inkjet printer 12, when the ink flying property of a part of the ejection port is lowered, the streak-like unevenness becomes conspicuous (easily manifested) in the logo and the background in the printed image.

On the other hand, in the image processing device 5 of the printing system 1, when the input image 7 is printed by the inkjet printer 12, the image areas 72 and 75 in which unevenness is easily noticeable in the printed image are specified as the target area by the target area specifying unit 51. Will be done. Then, the image conversion unit 52 generates a converted image (represented by a mixture of C, M, and Y) in which the color (K) of the target area of the input image 7 is converted into a pseudo color using C, M, and Y. Will be done. This makes it possible to increase the number of ink colors used for the target area in the inkjet printer 12. As a result, even when the ink flying property of a part of the ejection ports is lowered, the influence on the printed image can be alleviated, that is, unevenness in the target region can be suppressed in the printed image.

On the other hand, in the image region (non-target region) where unevenness is less noticeable in the printed image, a converted image is generated without changing the color of the input image 7. Therefore, printing can be performed faithfully to the input image 7. In the printing system 1, it is possible to form a preferable printed image by determining the necessity of pseudo-coloring according to the type (attribute) of the image in each image area. For two-dimensional codes such as barcodes and QR codes (registered trademarks), and image areas such as micro characters and thin lines (ruled lines), if printing (mixing) with inks of multiple colors, high registration accuracy is achieved. Desired. Therefore, also from such a viewpoint, it is preferable to print these image regions faithfully with only K ink faithfully to the input image 7.

In the image processing device 5, the area attribute data 82 indicating the attributes of each image area in the input image 7 is input to the target area specifying unit 51. Thereby, the target area can be easily specified based on the area attribute data 82. In the converted image, the density of K in the image area 72 showing the logo is made larger than 0%, so that the reproducibility of the original color of the image area 72 (the color in the input image 7) in the printed image can be improved. can. By setting the density of K in the image region 75 showing the background to 0% in the converted image, unevenness can be further made less noticeable in the image region 75 of the printed image. The image conversion unit 52 can appropriately determine the pseudo color of the target area in the converted image based on the type (attribute) of the image of the target area.

FIG. 5 is a block diagram showing another example of the image processing apparatus. In the image processing device 5 of FIG. 5, a learning unit 53 is added as compared with the image processing device 5 of FIG. Further, the processing of the target area specifying unit 51a is different from that of the target area specifying unit 51 of FIG.

The target area specifying unit 51a includes a trained model generated by a model generation process (learning) described later. Thereby, it is possible to specify each image area of the unknown input image 7 and acquire the attribute of the image area without using the area attribute data 82. In the printing process by the printing system 1 including the image processing device 5 of FIG. 5, the input image data 81 to which the area attribute data 82 is not attached is prepared (FIG. 3: step S11). In the target area specifying unit 51a, a plurality of image areas of the input image 7 are specified together with their attributes, and among the plurality of image areas, an image area having a predetermined attribute (designated attribute) is extracted as a candidate for the target area. NS. Then, among the candidates for the target area, the image area having C0%, M0%, Y0%, and K100% is specified as the target area (step S12).

The image conversion unit 52 generates a converted image in which the color of the target area of the input image 7 is converted into a pseudo color (step S13). At this time, the pseudo color of the target area is determined based on the attribute specified by the target area specifying unit 51a, that is, the type of the image. Then, by performing a shading process on the image of each color component of the converted image, a halftone dot image of the color component is generated (step S14), and the inkjet printer 12 prints the halftone dot image of each color component. Is performed (step S15).

Subsequently, a model generation process for generating the trained model of the target area specifying unit 51a will be described. In the model generation process, first, a teacher data group 80 including a plurality of image data sets is prepared and input to the learning unit 53. Each image data set is the same as the input image data set 8 of FIG. 2, and includes image data indicating an image for learning and area attribute data indicating the position, size, and attributes of each image area of the image. In the learning unit 53, the combination of each image area of the image for learning indicated by each image data set and the attribute associated with the image area is treated as the teacher data. Then, a trained model (a part of the target area specifying unit 51a) is generated by training a classifier using a neural network or the like using a plurality of teacher data indicated by the teacher data group 80. The trained model can identify each image area of the unknown input image 7 and acquire the attributes of the image area.

As described above, in the teacher data group 80, the combination of each image area of the image for learning and the attribute associated with the image area is treated as teacher data. Further, in this processing example, the image area of the specified attribute is extracted as a candidate of the target area in which unevenness is easily noticeable in the printed image. Therefore, the teacher data group 80 shows an image to which one label indicating an attribute (designated attribute) in which unevenness is easily noticeable in the printed image is given, and an attribute (attribute other than the designated attribute) in which unevenness is less noticeable in the printed image. It can be regarded as including the image with the label of.

As described above, also in the image processing device 5 of FIG. 5, the target area is specified from the input image 7 by the target area specifying unit 51a, and the color of the target area of the input image is converted into a pseudo color by the image conversion unit 52. The converted image is generated. As a result, unevenness in the target area can be suppressed in the printed image. Further, since a part of the target area specifying unit 51a is constructed by learning using the teacher data group 80, the target area can be easily specified from the input image 7. Further, the image conversion unit 52 can appropriately determine the pseudo color of the target area in the converted image based on the type of the image in the target area.

In the above-mentioned processing by the target area specifying unit 51a, after extracting the image area that is a candidate of the target area, the target area is specified based on the density of each color component of the image area. The target area may be directly specified from the input image without extracting the area candidates. In this case, the attribute of each image area is whether or not the unevenness is easily noticeable in the printed image. Therefore, the generation of the trained model includes an image with one label indicating that unevenness is easily noticeable in the printed image and an image with another label indicating that unevenness is less noticeable in the printed image. The teacher data group 80 is prepared. Then, by learning using the teacher data group 80, a trained model that is the entire target area specifying unit 51a is constructed. The target area specifying unit 51a can directly specify the attribute of each image area of the unknown input image 7, that is, whether or not the image area is the target area. As described above, in the image processing device 5 of FIG. 5, at least a part of the target area specifying unit 51a may be constructed by learning using the teacher data group 80.

Various modifications are possible with the image processing device 5 and the image processing method.

The target region may be a region in which the concentration fluctuates slightly. For example, in the image region where only K is used in the input image 7, the density of K changes slightly or gradually when the density of K is a predetermined value or more (for example, 60% or more, preferably 80% or more). Even in the case of (gradation), unevenness may be easily noticeable in the image area in the printed image. Therefore, the target region specified by the target region specifying units 51, 51a is a region in which the concentration is approximately constant above a predetermined value (including a region in which the concentration is strictly constant) or a region in which the concentration changes stepwise. It is preferable to have. When the density of K fluctuates in the target region, the density of a plurality of color components at the position in the converted image is also changed according to the density of K at each position.

In the above embodiment, in the target region of C0%, M0%, Y0%, and K100% in the input image 7, the colors of the target region are changed by increasing the concentrations of C, M, and Y to be larger than 0%. Although it is converted into a pseudo color, the color of the target area may be converted into a pseudo color by, for example, C10%, M0%, Y0%, and K100%. In this case, not only the K ink but also the C ink is used for the target area of the printed image, so that unevenness can be suppressed. In addition, the density of the target area of the printed image can be further increased.

Further, in a typical ink, when compared with C, M, Y, and K, the brightness of K is the lowest, so that the image region of C0%, M0%, Y0%, and K100% is specified as the target region. However, depending on the design of the image processing apparatus 5, for example, an image region having C100%, M0%, Y0%, and K0% may be specified as the target region. In this case, for example, by setting C70%, M10%, Y10%, and K10%, the color of the target area is converted into a pseudo color, and the inks of C, M, Y, and K are used for the target area of the printed image. Will be done. As described above, in the preferred image processing apparatus 5, the region where the density of three color components of C, M, Y, and K is 0%, that is, the density of only one color component is higher than 0%. A large region is specified as a target region, and the color of the target region is converted into a pseudo color by increasing the density of two or more color components in the target region to more than 0%.

Depending on the design of the image processing device 5, a region in which the density of only two color components of C, M, Y, and K is greater than 0% is specified as a target region, and the density of three or more color components is set to 0. The color of the target area may be converted into a pseudo color while being larger than%. In one example, an image region of C50%, M0%, Y0%, and K50% is specified as a target region, and the color of the target region is converted to C60%, M10%, Y10%, and K20%. Further, a region in which the density of only three color components of C, M, Y, and K is higher than 0% is specified as the target region, and the target region is defined while the density of the four color components is higher than 0%. The color of may be converted to a pseudo color.

As described above, in the image processing apparatus 5, the target area in which the density of one color component is 0% and the unevenness is easily noticeable in the printed image when printed by the inkjet printer 12 is from the input image 7. Be identified. Then, in the target area of the input image 7, the color of the target area of the input image 7 is converted into a pseudo color by changing the density of a plurality of color components while increasing the density of the one color component to more than 0%. The converted image is generated. As a result, when the converted image is printed by the inkjet printer 12, the number of ink colors used for the target area of the printed image can be increased as compared with the input image 7, and unevenness in the target area of the printed image can be increased. It can be suppressed.

As described above, when a region in which the density of only two or three color components of C, M, Y, and K is greater than 0% is specified as the target region, one processing example of the image conversion unit 52 Then, the color component having the highest density in the target area of the input image 7 is set as the color component of interest, and the density of the color component of interest is made larger than 0% in the target area of the converted image. This makes it possible to improve the reproducibility of the original color of the target area (the color in the input image 7) in the printed image. Further, in another processing example of the image conversion unit 52, the density of the color component of interest is set to 0% in the target region of the converted image. As a result, when the brightness of the ink of the color component of interest is low, unevenness can be made less noticeable in the printed image. When a region in which the density of only one color component is greater than 0% is specified as the target region, the color component is the color component of interest.

The attribute of each image area indicated by the area attribute data 82 may indicate whether or not the image area is a target area (whether or not the unevenness is easily noticeable in the printed image). In this case, for example, in a photographic image, it is possible to easily specify a region having a substantially constant density as a target region, such as a region showing a night sky, and another region as a non-target region.

In the inkjet printer 12 of FIG. 1, so-called single-pass printing is performed in which printing on the base material 9 is completed when the base material 9 passes under the ejection portion 2 once. , So-called multi-pass printing may be performed. In this case, for example, the ejection portion 2 moves intermittently in the X direction with respect to the substrate 9, and the substrate 9 passes under the ejection portion 2 a plurality of times to complete printing on the substrate 9. The movement of the base material 9 with respect to the discharge portion 2 may be relative.

The base material 9 does not necessarily have to be a continuous sheet, and may be cut to a predetermined length (cut paper or the like). Further, the base material 9 may be other than paper and film such as a thin metal plate.

The above-described embodiment and the configurations in each modification may be appropriately combined as long as they do not conflict with each other.

1 Printing system 5 Image processing device 6 Shading part 7 Input image 9 Base material 12 Inkjet printer 51, 51a Target area identification part 52 Image conversion part 71-75 Image area 80 Teacher data group 82 Area attribute data S11 to S15 Step

Claims (18)

An image processing device that processes input images for inkjet printers.
In the input image, the color at each position is represented by the densities of a plurality of color components, the density of one color component is 0%, and the inks of a plurality of colors corresponding to the plurality of color components are inks. A target area specifying unit that identifies a target area in which unevenness is likely to be noticeable in a printed image when printed by an inkjet printer using the above input image, and a target area specifying unit.
By changing the densities of the plurality of color components while increasing the density of the one color component to more than 0% in the target area of the input image, the color of the target area of the input image becomes a pseudo color. An image conversion unit that generates a converted converted image, and
An image processing device comprising. The image processing apparatus according to claim 1.
An image processing apparatus characterized in that the target area specified by the target area specifying unit is a region in which the density is substantially constant or the density changes stepwise. The image processing apparatus according to claim 1 or 2.
An image processing apparatus characterized in that the target area specified by the target area specifying unit is a region in which a single color component is used. The image processing apparatus according to any one of claims 1 to 3.
Area attribute data indicating the attributes of each area in the input image is input to the target area identification unit, and the area attribute data is input to the target area identification unit.
An image processing device characterized in that the target area specifying unit identifies the target area based on the area attribute data. The image processing apparatus according to any one of claims 1 to 3.
At least a part of the target area identification part is constructed by learning using the teacher data group, and is constructed.
An image in which the teacher data group includes an image to which one label indicating an attribute in which unevenness is easily noticeable is given in a printed image and an image to which another label is given to show an attribute in which unevenness is less noticeable. Processing equipment. The image processing apparatus according to any one of claims 1 to 5.
An image processing apparatus characterized in that the color component having the maximum density in the target region of the input image is a attention color component, and the density of the attention color component in the target region of the converted image is larger than 0%. The image processing apparatus according to any one of claims 1 to 5.
An image processing apparatus characterized in that the color component having the maximum density in the target region of the input image is the attention color component, and the density of the attention color component in the target region of the converted image is 0%. The image processing apparatus according to any one of claims 1 to 7.
An image processing device characterized in that the image conversion unit determines the pseudo color based on the type of an image in the target area. A halftone dot image generator that generates halftone dot images used in inkjet printers.
The image processing apparatus according to any one of claims 1 to 8.
A shading portion that generates a halftone dot image of each color component by performing a shading process on the image of each color component of the converted image generated by the image processing apparatus.
A halftone dot image generator, which comprises. It's a printing system
The halftone dot image generator according to claim 9 and
An inkjet printer that prints on a substrate by an inkjet method based on a halftone dot image of each color component generated by the halftone dot image generator.
A printing system characterized by being equipped with. An image processing method that processes an input image for an inkjet printer.
a) In the input image, the color at each position is represented by the densities of a plurality of color components, which is a region where the density of one color component is 0%, and a plurality of colors corresponding to the plurality of color components. A step of identifying a target area in which unevenness is easily noticeable in a printed image when printed by an inkjet printer using the ink of the above from the input image, and
b) In the target area of the input image, the color of the target area of the input image is simulated by changing the density of the plurality of color components while increasing the density of the one color component to more than 0%. The process of generating a converted image converted to color,
An image processing method comprising. The image processing method according to claim 11.
The image processing method, wherein the target region specified in the step a) is a region in which the density is substantially constant or the density changes stepwise. The image processing method according to claim 11 or 12.
The image processing method, wherein the target region specified in the step a) is a region in which a single color component is used. The image processing method according to any one of claims 11 to 13.
Area attribute data indicating the attributes of each area in the input image is prepared.
The image processing method, characterized in that, in the step b), the target area is specified based on the area attribute data. The image processing method according to any one of claims 11 to 13.
In the step a), the target area is specified by the target area identification unit, and the target area is specified.
At least a part of the target area identification part is constructed by learning using the teacher data group, and is constructed.
An image in which the teacher data group includes an image to which one label indicating an attribute in which unevenness is easily noticeable is given in a printed image and an image to which another label is given to show an attribute in which unevenness is less noticeable. Processing method. The image processing method according to any one of claims 11 to 15.
An image processing method characterized in that the color component having the maximum density in the target region of the input image is a attention color component, and the density of the attention color component in the target region of the converted image is larger than 0%. The image processing method according to any one of claims 11 to 15.
An image processing method characterized in that the color component having the maximum density in the target region of the input image is the attention color component, and the density of the attention color component in the target region of the converted image is 0%. The image processing method according to any one of claims 11 to 17.
The image processing method, characterized in that, in the step b), the pseudo color is determined based on the type of the image in the target area.

Images