JP2005223463A - Proof image generating apparatus, and proof image generating method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for easily generating a proof image of a dot image. <P>SOLUTION: A hard disk 104 of a proof image generating apparatus realized by a computer stores data 7 and a threshold matrix 81 of an original image, and a three-dimensional LUT generating section 21 uses the threshold matrix 81 to produce a plurality of binary screen tint images respectively corresponding to gradation levels which the original image converted to have a proof purpose resolution can take. The plurality of screen tint images are brought into images with a low resolution in matching with the proof purpose resolution, a plurality of screen tint images of multi-gradation are generated, and a three-dimensional LUT is generated on the basis of the plurality of multi-gradation images. A proof image generating section 22 references the three-dimensional LUT in a proof purpose unit region of the original image on the basis of two-dimensional coordinates and a gradation level of each pixel to acquire the gradation level of each pixel of the proof image. Thus, the proof image with multi-gradation can easily be generated without the need for bringing the original image to have a high resolution and to produce a dot image. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a technique for generating a proof image of a halftone image representing a multi-tone original image.

  Conventionally, when a printed product of a binary halftone image obtained by halftoning (also called halftone processing) of a multi-tone image is created, the color of the halftone image after printing, moire, etc. In order to calibrate the image by estimating the state of the image, a proof image (that is, an image used for proofreading, hereinafter referred to as a “proof image”) is generated, and a printed proof that is the printed matter is created. Is done.

  When creating a print proof, first, data of an original image to be halftone (hereinafter referred to as “original image”) is converted into halftone dot area ratio data, and the halftone dot area ratio data is converted into binary bitmap data. The halftone image is generated by developing. Then, a multi-gradation proof image is generated by reducing the number of pixels of the halftone image as a whole and a multi-gradation proof image is generated (for example, as a method of generating a multi-gradation proof image from a halftone image, Patent Document 1 Printed by a density gradation type color printing apparatus (also called a continuous gradation method, including one that expresses the gradation of each pixel by the dither method), and creates a print proof Is done.

In Patent Document 2, a threshold value is prepared for each of a plurality of small sections obtained by dividing each pixel of the original image, and the total of binary comparison results between the gradation level of each pixel and the corresponding threshold values of the plurality of small sections is prepared. A technique for generating a multi-tone halftone image by assigning multi-tone levels to each pixel according to the value is disclosed.
US Pat. No. 5,854,883 JP 11-98356 A

  By the way, in the print proof creation method, since a proof image is generated after generating a high-resolution halftone image having a large data capacity, a complicated process requiring a large memory capacity is required, and an expensive image generation apparatus is required. If it is not used, it takes a long time to process.

  The present invention has been made in view of the above problems, and provides a technique for easily generating a proof image of a halftone image representing a multi-tone original image without generating a high-resolution halftone image. It is aimed.

  The invention according to claim 1 is a proof image generation device that generates a proof image of a halftone image representing a multi-tone original image, the original image data and the high-resolution original image The threshold value matrix for each of the gradation levels that can be taken by the pixels of the original image converted to the resolution for proofing, and a storage unit for storing a threshold value matrix used when halftone dots are formed for each predetermined unit area Is used to generate a binary flat mesh image of the unit area having a uniform gradation level, and the binary flat mesh image is substantially reduced in resolution according to the resolution for the proof so as to obtain a multi-tone image. In each unit area of the original image converted into the proof resolution, means for generating a three-dimensional table with two-dimensional coordinates and gradation levels as parameters by further generating a flat mesh image, By referring to the 3-dimensional table based on two-dimensional coordinates and the gradation level of the unit, and means for generating a multi-tone proof images.

  The invention according to claim 2 is the proof image generation device according to claim 1, wherein the binary flat mesh image is a row a column (a is an integer of 2 or more), and the binary image The flat halftone image is (n1 / m1) times in the row direction (n1, m1 are relatively prime natural numbers, and n1 <m1), and (n2 / m2) times in the column direction (n2, m2 are prime numbers that are relatively prime), and , N2 <m2), the means for generating the three-dimensional table expands the binary flat screen image by n1 times in the row direction and n2 times in the column direction, and a and m1 And the greatest common divisor b2 between a and m2 are calculated, respectively, and (m1 / b1) enlarged flat mesh images in the row direction and (m2 / b1) in the column direction. b2) A new flat net image arranged one by one is generated, and representative values of element values of m1 rows and m2 columns are generated from the new flat net image as elements. Generating a tint image of the multi-tone having a.

  A third aspect of the present invention is the proof image generating apparatus according to the first or second aspect, wherein high-frequency components are removed before the resolution of the binary flat screen image is reduced.

  The invention according to claim 4 is the proof image generation device according to any one of claims 1 to 3, wherein the proof image generated by the means for generating the proof image is corrected depending on the proof printing device. Further, a correction means for performing is provided.

  The invention according to claim 5 is a proof image generation method for generating a proof image of a halftone image representing a multi-tone original image, the original image data and the high-resolution original image For each of the gradation levels that can be taken by the pixels of the original image converted to the resolution for proofing, a threshold matrix used for halftone dot conversion for each predetermined unit area is prepared. A binary flat image of the unit area having a uniform gradation level is generated, and the binary flat image is reduced to a resolution substantially corresponding to the resolution for the proof to obtain a multi-tone flat image. In each unit region of the original image converted into the resolution for proof, a step of generating a three-dimensional table with two-dimensional coordinates and gradation levels as parameters by further generating a halftone image By referring to the 3-dimensional table based on two-dimensional coordinates and the gradation level, and a step of generating a multi-tone proof images.

  The invention according to claim 6 is a program for causing a computer to generate a proof image of a halftone image representing a multi-tone original image, and the execution of the program by the computer causes the computer to And a step of preparing a threshold matrix used when the original image with high resolution is converted into halftone dots for each predetermined unit area, and pixels of the original image converted to the proof resolution can be taken. For each gradation level, a binary flat image of the unit area having a uniform gradation level is generated using the threshold value matrix, and the binary flat image is substantially matched to the proof resolution. Generating a three-dimensional table using two-dimensional coordinates and gradation levels as parameters by further reducing the resolution and further generating a multi-tone flat mesh image; A step of generating a multi-gradation proof image by referring to the three-dimensional table based on the two-dimensional coordinates and gradation levels of each pixel in each unit area of the original image converted to the proof resolution. And execute.

  According to the first to sixth aspects of the present invention, it is possible to easily generate a proof image without generating a high-resolution halftone dot image.

  In the invention of claim 2, a multi-tone flat halftone image can be generated appropriately and efficiently.

  In the invention of claim 3, aliasing noise can be suppressed.

  In the invention of claim 4, the proof image printed by the proof printing apparatus can be further approximated to a halftone dot image.

  FIG. 1 is a diagram for explaining an outline of processing for creating a printed image of a halftone image representing a multi-tone original image. When creating a printed image of a halftone image, a proof image is first generated from a multi-tone original image based on various parameters for generating a halftone image, and a proof print proof is printed. Subsequently, the various parameters are appropriately modified based on the print proof, and the print proof is printed again as necessary. Thereafter, a halftone image is generated from the original image, and a printed matter of the halftone image is created. Hereinafter, for the convenience of description, an outline of processing for generating a halftone image will be described first, and then details of processing for generating a proof image will be described.

  In the following description, the resolution of an image means the number of pixels, high resolution means increasing the number of pixels of the image to be processed, and low resolution means the number of pixels of the image. It means to reduce. In addition, the degree of density (for example, expressed in dpi (dot per inch)) of a manifested image such as an image printed on a print medium or an image displayed on a display is expressed as recording resolution. To do.

  When generating a halftone image, as indicated by solid line arrows 911, 912, and 913 in FIG. 1, the original image is C (cyan), M (magenta), Y (yellow), and K (black). The resolution is converted for each color plate by a predetermined method to increase the number of pixels (that is, the resolution is increased), and an original image having the same resolution as the halftone image to be generated is acquired. The original image of each color plate of CMYK having a higher resolution is halftoned by being compared with a threshold matrix (also referred to as SPM (Screen Pattern Memory) data) described later, and a halftone image of each color plate is generated. .

  FIG. 2 is a diagram for explaining a state in which the original image 71 having a high resolution is formed into halftone dots, and shows a part of the image 71 and a threshold matrix 81. As shown in FIG. 2, a plurality of halftone dot unit areas 711 serving as a halftone dot unit divided into a plurality of areas of the same size are set in the original image 71 with high resolution. Here, a plurality of unit areas are set in the original image before resolution conversion corresponding to the plurality of halftone dot unit areas 711, and are included in each unit area in the original image when the resolution of the original image is increased. Similarly, the number of pixels to be displayed is increased so that the unit area becomes a halftone unit area 711. The threshold value matrix 81 has a plurality of threshold values corresponding to a plurality of pixels included in each halftone dot unit area 711 as element values. In FIG. 2, each element is regarded as a pixel and the threshold value matrix 81 is illustrated. Show.

  When halftoning the high-resolution original image 71, conceptually, the halftone dot unit areas 711 and the threshold matrix 81 of the original image 71 are superposed to each pixel of the halftone dot unit area 711. By comparing the gradation level (pixel value) with the corresponding element value (threshold value) of the threshold matrix 81, the gradation level of the position (coordinate) of the pixel in the binary halftone image is determined. Therefore, if the gradation level of the image 71 is uniform, for example, the gradation level “1” is given to the pixel of the coordinate for which the threshold value smaller than the gradation level is set in the threshold value matrix 81. Then, the gradation level “0” is given to the remaining pixels, and macroscopically uniform halftone dots are generated. Actually, since the original image 71 with high resolution has light and shade (that is, portions of various gradation levels), the halftone dot state changes in the halftone dot unit region 711 according to the lightness and darkness of the image 71. A point image is generated. Note that the processing of converting the original image 71 having a high resolution using the threshold matrix 81 into halftone dots for each halftone unit area 711 converts the original image into halftone dot area ratio data, and distributes the halftone dot area ratio distribution. This is substantially the same as the process of generating binary bitmap data that is a halftone dot that satisfies the above.

  When a halftone image (for example, an image represented by CMYK four-version 1-bit TIFF format data) is generated from the original image, the halftone image of each color plate of CMYK is displayed on one print medium in the image recording apparatus. To produce a color dot image print. When each element of the threshold matrix 81 is regarded as a pixel as shown in FIG. 2, a set of threshold matrixes 81 (hereinafter referred to as “corresponding to the original image”) in which the threshold matrix 81 is arranged corresponding to the halftone dot unit region 711. It can be said that the “resolution” is the same resolution as the halftone image to be generated.

  Next, processing for generating a proof image will be described. When generating a proof image for calibration or confirmation before generating a halftone image, as shown by solid line arrows 914, 915, 916, and 917 in FIG. By generating a plurality of images in accordance with the proof resolution, a three-dimensional lookup table (hereinafter referred to as “three-dimensional LUT”) is generated. Converted to proof resolution. Then, based on the original image after conversion and the three-dimensional LUT, four color multi-tone proof images (each image is in 8-bit-TIFF format and corresponds to one 32-bit-TIFF image data. ) Is generated and a print proof is created. Hereinafter, a method for creating a print proof will be described in detail.

  FIG. 3 is a diagram showing the configuration of the print proof creation system 1. The printing proof creation system 1 includes a computer 11 and a printing device for printing proof (hereinafter referred to as “printer”) 12. The ink jet printer 12 receives a signal from the computer 11 and generates fine droplets from a plurality of nozzles. An image is recorded by discharging onto a print medium. The printer 12 may be a printing apparatus of another density gradation method (here, the ink jet method is also regarded as the density gradation method in pixel units).

  The computer 11 has a general computer system configuration in which a CPU 101 that performs various arithmetic processes, a ROM 102 that stores basic programs, and a RAM 103 that stores various information are connected to a bus line. The bus line further includes a fixed disk 104 for storing information, a display 105 for displaying various information, a keyboard 106a and a mouse 106b for receiving input from an operator, an optical disk, a magnetic disk, a magneto-optical disk, and the like. A reading device 107 that reads information from the recording medium 91, and a communication unit 108 connected to a computer for generating an original image, a device for recording halftone images, and the like appropriately via an interface (I / F), etc. Connected.

  The computer 11 reads the program 92 from the recording medium 91 via the reading device 107 in advance and stores it in the fixed disk 104. When the program 92 is copied to the RAM 103 and the CPU 101 executes arithmetic processing according to the program in the RAM 103 (that is, when the computer executes the program), the computer 11 expresses a multi-tone original image. An operation as a proof image generation device for generating a proof image of a halftone image is performed.

  FIG. 4 is a block diagram illustrating a functional configuration realized by the CPU 101, the ROM 102, the RAM 103, the fixed disk 104, and the like when the CPU 101 operates in accordance with the program 92. In FIG. Indicates the function to be realized. In addition, the function of the calculating part 20 may be implement | achieved by a dedicated electrical circuit, and an electrical circuit may be used partially.

  FIG. 5 is a diagram showing a flow of processing in which the computer 11 generates a proof image. Hereinafter, a process for generating a proof image of one color plate from one original image of the four color plates of CMYK will be mainly described with reference to FIG. 4, but the same applies to proof images of other color plates. Is generated. In the following description, the original image is expressed by 256 gradations of 0 to 255, but the number of gradations of the original image is not limited to this.

  In the computer 11, first, original image data (hereinafter referred to as “original image data”) 7 and a threshold matrix 81 are obtained from another computer via the communication unit 108 and stored in the fixed disk 104 and prepared. (Step S11). Further, print condition data 93 to be described later used for color matching when the generated proof image is printed by the printer 12 is further acquired and stored in the fixed disk 104.

  Subsequently, the threshold matrix 81 is output to the three-dimensional LUT generation unit 21, and a plurality of images matching the resolution for proofing the original image is generated using the threshold matrix 81, thereby generating a three-dimensional LUT ( Step S12). Here, the resolution for proofing is determined in advance based on the ratio between the recording resolution of the image recording apparatus used for creating the printed matter of the halftone image and the recording resolution of the printer 12. For example, when the recording resolution of the image recording apparatus is 2400 dpi and the recording resolution of the printer 12 is 720 dpi, the resolution for proofing is determined as (3/10) times the resolution of the halftone image. The three-dimensional LUT generation unit 21 generates a binary flat mesh image indicating a uniform gradation level for each gradation level from 0 to 255 using the threshold matrix 81, and sets the resolution for proofing the original image. In addition, the resolution of this flat halftone image is substantially reduced to a multi-tone flat halftone image. Then, 256 multi-gradation flat halftone images are stacked in order of gradation levels to generate a three-dimensional LUT.

  In the following description, a binary flat screen image is converted into an image in which an integer number of pixels are arranged in the row and column directions due to low resolution, and this image is treated as one multi-tone flat screen image of a three-dimensional LUT. However, a case where the number of pixels in the row and column directions of the binary flat mesh image does not become an integer due to the reduction in resolution will be described separately.

  FIG. 6 is a diagram showing a multi-tone flat halftone image 84 that is one layer of the three-dimensional LUT generated by the three-dimensional LUT generation unit 21. The multi-tone flat halftone image 84 in FIG. 6 is for the gradation level 128, and elements arranged in six rows from the first row to the sixth row and six columns from the A column to the F column are shown. Have. In the multi-grayscale flat mesh image 84 of FIG. 6, the A column or the F column, or the first row or the first column, centering on the element where the D column intersects the fourth row (hereinafter referred to as the element (D, 4)). As the value approaches 7 rows, the element value (that is, the pixel value) gradually decreases, and the distribution of the element value becomes a mountain shape in any of the two arrangement directions. As described above, in the present embodiment, a flat halftone image is called including an image showing only one dot of a halftone dot. Details of processing for generating a three-dimensional LUT using the threshold matrix 81 will be described after the description of the proof image generation processing.

  When the three-dimensional LUT is generated, the original image data 7 is output from the fixed disk 104 to the proof image generation unit 22, and the original image is enlarged by a predetermined method and converted to a proof resolution (step S13). . At this time, each unit area set in the original image is also converted to a proof resolution (hereinafter, the converted unit area is referred to as a “proof unit area”). It is assumed that the original image converted to the proof resolution is also expressed in 256 gradations from 0 to 255.

  Subsequently, the proof image generation unit 22 refers to the three-dimensional LUT based on the two-dimensional coordinates and the gradation level of each pixel in each proof unit area of the converted original image, so that a multi-gradation proof image is obtained. Is generated (step S14). For example, when each pixel is specified by coordinates (address) in the proof unit area 712 of the original image shown in FIG. 7, if the gradation level of the pixel at coordinates (A, 1) is 128, a three-dimensional LUT 6 is selected, and the element value 0 of the element (A, 1) is set to the gradation level of the corresponding pixel in the proof image. When the gradation level of the pixel at the coordinate (B, 3) in the proof unit area 712 is 128, the element value 191 of the element (B, 3) in the multi-tone flat halftone image 84 is the value in the proof image. The gradation level of the corresponding pixel is set. Similarly, the gradation level of other pixels in the proof unit area 712 is specified, and the element value of the same coordinate of the multi-tone flat halftone image of that gradation level is set as the gradation level of the corresponding pixel of the proof image. The

  As described above, the gradation level of each pixel of the proof image is obtained by referring to one multi-gradation flat halftone image of the three-dimensional LUT based on the two-dimensional coordinates of the pixel of the corresponding proof unit area 712. Therefore, if the gradation levels of the pixels in the proof unit area 712 are uniform, the values of each multi-gradation flat halftone image of the three-dimensional LUT increase stepwise toward the center as described above. Since the distribution of the element values is of the (mountain shape) type (of course, the distribution of the element values is not limited to this), the proof unit area 712 also has a gradation level in the generated proof image. Yamagata distribution. Actually, the gradation level of the pixels in the proof unit area 712 has a gray level, and since this gray level is reflected in the proof image via the three-dimensional LUT, the halftone image which is the final printed matter (see FIG. 1). In this case, a proof image that reproduces moire or the like generated by interference between the period of the halftone dot unit area 711 and the periodicity included in the original image is generated. Further, since a multi-gradation flat halftone image corresponding to the gradation level of each pixel in the proof unit area 712 is selected from the three-dimensional LUT, the hue of the original image is not impaired by the proof image.

  When a CYMK proof image is generated by performing the processing of steps S11 to S14 for each color plate, the proof image data is output to the proof image correction unit 23. Print condition data 93 is input from the fixed disk 104 to the proof image correction unit 23, and a process for correcting the proof image based on the print condition data 93 is performed (step S15).

  Specifically, the print condition data 93 indicates information for correcting the color of the proof image in accordance with the print medium used by the printer 12 and the type of ink, and the proof image correction unit 23 uses the CYMK. Color correction (so-called color matching) depending on the printer 12 is performed on the proof image of each color plate, and the proof image after correction is binarized by an error diffusion method and output to the printer 12. A print proof is created.

  At this time, since the color matching by the proof image correcting unit 23 is performed by shifting the gradation level of the pixels in each color plate, it is printed by the printer 12 while maintaining the gradation level distribution of the proof image. The proof image can be further approximated to a halftone image. The print condition data 93 may be color matching information based on an image recording apparatus used when printing a halftone image, for example.

  Next, details of the process (step S12) for generating the three-dimensional LUT will be described. FIG. 8 is a diagram showing a flow of processing in which the three-dimensional LUT generation unit 21 generates a three-dimensional LUT, and FIG. 9 is a diagram showing a threshold matrix 81. In FIG. 9, the element value is illustrated as a gradation level, and the element value is smaller as the width of the parallel diagonal line is narrower. In practice, a threshold value matrix having a distribution of element values having a plurality of local maximum points and local minimum points as shown in FIG. 9 is used. However, in the present embodiment, one local maximum point is used to simplify the description. In the following description, the distribution of element values having local minimum points (that is, the distribution corresponding to one halftone dot in FIG. 9) is used as the threshold value matrix 81.

  In the three-dimensional LUT generation unit 21, first, a halftone dot unit area 711 having a uniform gradation level is prepared for each of gradation levels 0 to 255 that can be taken by pixels of the original image converted to the resolution for proofing. The halftone dot unit area 711 is shaded using the threshold matrix 81 to generate a binary flat halftone image (step S21).

  For example, when the halftone dot unit area 711 having the gradation level 128 is prepared, the gradation level “1” is set to, for example, the pixel of the coordinate in which the threshold value smaller than 128 is set in the threshold value matrix 81. 10. The gradation level “0” is assigned to the remaining pixels, and FIG. As shown in A, a binary flat image 82a relating to the gradation level 128 is generated (however, FIG. 10.A shows a flat image corresponding to the actual threshold matrix of FIG. 9). Further, when an image of gradation level 0 is prepared, gradation level “0” is assigned to all pixels, and FIG. As shown in B, a binary flat halftone image 82b relating to the gradation level 0 is generated. In this way, 256 flat halftone images respectively associated with 256 gradation levels are generated. FIG. C to FIG. E denotes a flat halftone image 82c having a gradation level 32, a flat halftone image 82d having a gradation level 223, and a flat halftone image 82e having a gradation level 225, respectively. Note that it is not always necessary to prepare the halftone dot unit area 711 having a uniform gradation level for each of the gradation levels from 0 to 255, and the threshold value matrix 81 is compared with a single threshold value to obtain a binary value. A flat mesh image may be generated.

  Subsequently, in order to reduce the resolution of the binary flat halftone image substantially in accordance with the resolution for proofing of the original image, the resolution for proofing and the resolution of the halftone dot image as the final printed matter (the original image of the threshold matrix 81). The resolution conversion magnification is obtained based on the resolution of the corresponding array (step S22). Here, in order to simplify the explanation, it is assumed that the resolution conversion magnification is obtained as (1/2). FIG. 8 shows the case where the magnification is (n / m). Subsequently, each flat screen image is enlarged using the numerator value 1 of the resolution conversion magnification as a magnification (step S23).

  FIG. 11 is a diagram showing a flat mesh image 82 relating to the gradation level 128. Pixels (elements) arranged in 12 rows from the first row to the twelfth row and 12 columns from the A column to the L column are shown. Show. Here, since the value of the numerator of the magnification for resolution conversion is 1, the flat mesh image 82 in FIG. 11 is handled as the enlarged flat mesh image 82 as it is. Note that when the flat mesh image 82 is actually enlarged, for example, a bicubic method, a bilinear method, or the like is used.

  In the three-dimensional LUT generation unit 21, the greatest common divisor 2 (the number of columns (number of columns) 12 (corresponding to “a” in FIG. 8) of the flat mesh image 82 before enlargement and the denominator value 2 of the resolution of the resolution conversion) (Corresponding to b in FIG. 8) is calculated (step S24). Subsequently, a flat mesh image 82 (that is, the flat mesh image 82 in FIG. 11) expanded by the number of the value 1 obtained by dividing the denominator value 2 of the resolution conversion magnification by the greatest common divisor 2 is displayed in each array direction. A new flat screen image is generated by arrangement (step S25). Here, the flat mesh image 82 of FIG. 11 is handled as a new flat mesh image 82 as it is.

  Then, element values of 2 rows and 2 columns having the denominator value 2 of the resolution conversion magnification as the number of rows and the number of columns are sequentially extracted from the new flat screen image 82, and have representative values of these element values as element values. A multi-tone flat mesh image is generated (step S26). For example, in the new flat net image 82 of FIG. 11, (A, 1), (A, 2), (B, 1), (B, 2) (hereinafter, ((A, 1) to (B, 2)) When the element value of () is extracted), the sum 0 of these element values is obtained as a representative value, and the element values of ((A, 7) to (B, 8)) are extracted. In this case, the sum 3 of element values is obtained as a representative value.

  In this way, the representative values of 36 element values of 2 rows and 2 columns included in the flat mesh image 82 of FIG. 11 are acquired, respectively, and as shown in FIG. 12, 36 representative values are used as element values. A 6-row, 6-column multi-tone flat mesh image 83 is generated appropriately and efficiently.

  When 256 multi-gradation flat halftone images 83 corresponding to gradation levels 0 to 255 are generated, the element values are changed in accordance with the gradation level range 0 to 255 in each flat halftone image 83. The For example, in the case of a flat halftone image 83 having a gradation level of 128 in FIG. 12, each element value is divided by 4 and then changed to a value obtained by multiplying by 255 to obtain a multi-gradation as shown in FIG. A flat mesh image 84 is generated. Conceptually, each of the 256 multi-gradation flat halftone images 84 is layered in the order of associated gradation levels, and a three-dimensional LUT is generated using two-dimensional coordinates and gradation levels as parameters (step). S27). Any method for generating the multi-gradation flat screen image 84 from the flat screen image 83 may be used as long as the distribution of element values is maintained.

  When the value obtained by dividing the denominator of the resolution conversion magnification by the greatest common divisor in step S25 is 1, an integer number of pixels in the binary and half-tone images in the row and column directions due to low resolution. Is converted into an image in a line, and this image is treated as a single multi-tone flat halftone image of a three-dimensional LUT, but the value obtained by dividing the denominator of the resolution conversion magnification by the greatest common divisor is 1. The case where this is not the case will be described later.

  As described above, in the proof image generation apparatus realized by the computer 11, a plurality of binary flat images respectively corresponding to a plurality of gradation levels are used by using the threshold value matrix 81 used when the original image is halftoned. A halftone image is generated, and each flat halftone image is multi-valued while being reduced in resolution according to the resolution for proofing. A three-dimensional LUT is generated by laminating a plurality of multi-gradation flat mesh images, and the three-dimensional LUT is referred to based on the coordinates and gradation levels in each pixel of the proof unit area of the original image. As a result, a proof image is generated. As a result, as in the conventional proof image generation method indicated by the dashed arrow in FIG. 1, a high resolution of the original image is generated to generate a halftone dot image, and a large amount of proof image is generated by reducing the resolution to a predetermined resolution. A proof image can be generated easily and at high speed without performing processing (a so-called 1-bit workflow). As a result, in the print proof creation system 1, even when an inexpensive printer 12 having a low recording resolution is used, an appropriate proof image that matches the printer 12 is generated, and a print proof that approximates a halftone dot printed matter. Can be created.

  Next, another example of processing for generating a three-dimensional LUT will be described. FIG. 13 is a diagram showing a flow of a three-dimensional LUT generation process according to another example, and shows a process performed between step S21 and step S22 of FIG.

  In a three-dimensional LUT generation process according to another example, when a binary flat screen image is generated (step S21), the three-dimensional LUT generation unit 21 performs gradations in successive pixel groups on each flat screen image. Low-pass filter processing is performed to remove high-frequency components of the waveform indicating the level change (step S31). Then, the processed halftone image is subjected to steps S22 to S26 to generate a multi-tone halftone image.

  In general, in the conventional proof image generation method shown in FIG. 1, aliasing noise caused by resolution conversion magnification when a halftone image is reduced in resolution (that is, sampling when resolution conversion is regarded as sampling) It is difficult to remove after conversion, and it is complicated to calculate aliasing noise during resolution conversion. Therefore, an expensive apparatus is required for processing in a short time. However, in the three-dimensional LUT generation processing according to another example, the high-frequency component is removed from the binary flat screen image before the resolution is reduced, so that the multi-tone flat screen image 83 is generated. Aliasing noise can be easily suppressed.

  Note that tone curve correction may be performed on a binary flat screen image as necessary. In step S31, a multi-tone flat halftone image is generated by low-pass filter processing. However, the processing in steps S22 to S26 can be similarly performed on a multi-tone flat halftone image. it can.

  In the proof image generation unit 22, a filtering process for removing high frequency components is applied to the original image before being converted to the proof resolution as necessary, and aliasing noise generated in the proof image is suppressed. Good. Further, tone curve correction and filter processing may be performed on the proof image to correct the nonlinearity of resolution conversion.

  Next, the case where the value obtained by dividing the denominator of the resolution conversion magnification by the greatest common divisor in step S25 in FIG. In this case, the value of (a × n / m) does not become a natural number in the process of multiplying the binary flat net image of row a and column a by (n / m) (steps S22 to S26). For example, assuming that the resolution conversion magnification of a binary flat mesh image of 10 rows and 10 columns is obtained as (2/3) times in accordance with the resolution for proof (FIG. 8: step S22), the binary flat mesh image is obtained. The image is enlarged twice (step S23), and a flat mesh image 820a enlarged in 20 rows and 20 columns shown in FIG. 14 is generated. Note that the flat mesh image 820a in FIG. 14 is enlarged while maintaining the distribution of the pixel values (element values) of the original flat mesh image, and in FIG. Distribution is shown.

  Subsequently, the greatest common divisor 1 between the number of rows (number of columns) 10 of the original flat mesh image and the denominator value 3 of the resolution conversion magnification is calculated (step S24), and the denominator value 3 of the resolution conversion magnification is calculated. The enlarged flat mesh images 820a of FIG. 14 are arranged in the arrangement direction by the number 3 obtained by dividing by the greatest common divisor 1. As shown in FIG. 15, a new flat mesh image 820b of 60 rows and 60 columns is formed. It is generated (step S25). In the new flat net image 820b of FIG. 15, the distribution of element values of the original flat net image is arranged in (3 × 3) in each arrangement direction.

  Then, element values of 3 rows and 3 columns having the denominator value 3 of the resolution conversion magnification as the number of rows and the number of columns are sequentially extracted from the new flat mesh image 820b, and have representative values of these element values as element values. A multi-tone flat mesh image 830 of 20 rows and 20 columns is generated as shown in FIG. 16 (step S26). In FIG. 16, the original flat mesh image 820 is shown as a comparison target in an overlapping manner with a two-dot chain line.

  Similarly to the new flat mesh image 820b, the distribution of element values of the original flat mesh image 820 is arranged (3 × 3) in each arrangement direction in the 20-row 20-column multi-tone flat mesh image 830 shown in FIG. . Therefore, the distribution of one element value (distribution in the rectangular area 822 indicated by a broken line in FIG. 16) in the multi-tone flat mesh image 830 is the distribution of the element values of the original flat mesh image 820 (2 / 3) It can be said that it is doubled. A three-dimensional LUT is generated by stacking a plurality of multi-gradation flat mesh images 830 generated in this manner.

  Here, considering the process of creating a proof image using a three-dimensional LUT based on the multi-tone flat mesh image 830 of FIG. 16 (FIG. 5: step S14), the proof resolution shown in FIG. 17 is converted. When the gradation level of the original image 72 is uniform, one multi-gradation flat mesh image 830 is superimposed on (3 × 3) proof unit regions 721 in the original image 72. And the pixel value is replaced. That is, (3 × 3) proof unit areas 721 have pixels of 20 rows and 20 columns in total, and a multi-tone proof image is obtained by referring to the 3D LUT based on the 2D coordinates of each pixel. Is generated. Actually, since the original image 72 has light and shade, the multi-tone flat image 830 selected according to the gray level of each pixel is different, but the size of the array of elements of all the flat images 830 is the same. Therefore, each pixel in the (3 × 3) proof unit areas 721 is replaced with the corresponding element value of any one of the flat mesh images 830.

  At this time, the proof unit area 721 in the original image 72 corresponding to one rectangular area 822 of the multi-tone flat mesh image 830 includes a pixel group corresponding to approximately 6.7 rows and 6.7 columns. The arrangement of the pixel group corresponding to the original image matches the resolution for proof (that is, the number of pixels of the original image 72 converted to the resolution for proof). Therefore, it can be said that a multi-tone flat net image 830 in which the binary flat net image 820 of 10 rows and 10 columns is substantially reduced in resolution according to the resolution for proof is generated by the above processing. In practice, the element value on the boundary between two rectangular regions 822 adjacent to each other in the multi-tone flat mesh image 830 of FIG. 16 is a value reflecting the distribution of these rectangular regions 822. In the original image 72 of FIG. 17, in the (3 × 3) proof unit regions 721 corresponding to one multi-gradation flat mesh image 830, pixels are located between two adjacent proof unit regions 721. Shared.

  The processing (steps S22 to S26) in the three-dimensional LUT generation unit 21 can be generally shown as follows. When the resolution conversion magnification is calculated as (n / m) times (n and m are prime natural numbers and n <m) (step S22), a row a column (a is an integer of 2 or more) ) Is reduced in resolution by (n / m) times. For this operation, the flat image is first enlarged n times (step S23). Subsequently, the greatest common divisor b between a and m is calculated (step S24), and (m / b) enlarged flat mesh images are arranged in each arrangement direction, and the circularity (period) of the element values A new flat screen image in which the property is stored is generated (step S25). Then, a multi-tone flat net image having a representative value of element values in m rows and m columns as element values is generated from the new flat net image (step S26). The representative value of the element values in m rows and m columns is not necessarily the sum of the element values, and may be an average value, for example.

  In the above description, an example in which the resolution of a binary flat mesh image is reduced by the same magnification in the row direction and the column direction has been shown. However, the binary flat mesh image has different magnifications in the row direction and the column direction. Even when the resolution is reduced, the above method can be extended and used. According to FIG. 8, the resolution conversion magnification of a binary flat screen image of a row and a column (a is an integer of 2 or more) is (n1 / m1) times (n1 and m1 are relatively prime in the row direction). If it is obtained as a natural number and n1 <m1) and (n2 / m2) times (n2 and m2 are relatively prime natural numbers and n2 <m2) in the column direction (step S22), first, 2 The flat image of the value is enlarged n1 times in the row direction and n2 times in the column direction (step S23). Subsequently, the greatest common divisor b1 between a and m1 and the greatest common divisor b2 between a and m2 are respectively calculated (step S24), and the enlarged flat screen image is obtained in the row direction (m1 / b1) and (m2 / b2) are arranged in the column direction, and a new flat screen image is generated (step S25). Then, a multi-tone flat net image having the representative values of the element values of m1 rows and m2 columns as element values is generated from the new flat net image (step S26). Accordingly, even when the recording resolution of the printer 12 is different in two directions corresponding to the row direction and the column direction of the image (for example, 1440 [dpi] × 720 [dpi]), Therefore, it is possible to generate a multi-tone flat mesh image in accordance with the resolution for proof according to the above, and as a result, it is possible to appropriately generate a proof image.

  Next, another example of processing (FIG. 8: Steps S22 to S26) for generating a multi-tone flat net image from a binary flat net image in the three-dimensional LUT generation unit 21 will be described. FIG. 18 is a diagram illustrating a pixel group 821 in one column of a binary flat halftone image. In FIG. 18, the gradation level of pixels with parallel diagonal lines is “0”, and the gradation levels of the remaining pixels are “1”.

  For example, when a resolution of a flat screen image is converted to (3/10) times, first, each pixel is given a weight of the numerator value 3 of the resolution conversion magnification and counted from the pixel on the left side in FIG. The value obtained by adding the product of the gradation level of each of the first, second and third pixels and the weight 3 and the product of the gradation level of the fourth pixel and the weight 1 is (0 × 3 + 0). X3 + 0x3 + 1x1 = 1). That is, the product of the gradation level and the weight is added until the total weight becomes equal to the denominator of the resolution conversion magnification. Next, the product of the gradation level of the fourth pixel and the remaining weight 2, the product of the gradation level of each of the fifth and sixth pixels and weight 3, and the gradation level of the seventh pixel And the product of the weight 2 are obtained as (1 × 2 + 1 × 3 + 0 × 3 + 0 × 2 = 5). If the flat screen image is multiplied by (3/10) times in the row direction by the above calculation, it is then converted to (3/10) times in the column direction by the same calculation. When the total weight k is less than 10 in the last calculation of the row or column, the total product of the gradation level and the weight is multiplied by (10 / k). With the above processing, it is possible to convert the resolution of a flat screen image to (3/10) times while reducing the memory capacity to be used.

  In general, in the above-described method, when the resolution of a flat screen image is converted to (n / m) times, a process of assigning n weights to all pixels, and the gradation of pixels until the sum of the weights becomes m It can be said that it consists of a process of adding the product of level and weight. Note that these processes in the case of (k = m) include a process (step S23) of enlarging the flat screen image by n times when (m / b) is 1 in the process of step S25, and a new process. This can be said to be equivalent to a process (step S26) for generating a multi-tone flat net image having a representative value of element values of m rows and m columns as an element value from the flat net image. Of course, also in the above method, the resolution conversion magnification may be different between the row direction and the column direction.

  In addition, the pixel value of a specific coordinate in a new flat net image in which a flat net image of a row and a column is enlarged n times and (m / b) is arranged in the row and column directions is multiplied by n times. In the enlarged flat net image, since the coordinate value of this specific coordinate is equal to the value of the pixel with the remainder obtained by dividing the coordinate value by (n × a), in practice, in the row and column directions (m / b) It is possible to realize an operation equivalent to an operation for the new flat net image without generating a new flat net image arranged one by one. As a result, it is possible to generate a multi-tone flat halftone image while reducing the memory capacity to be used.

  Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications can be made.

  In the above embodiment, unlike an actual threshold matrix (see FIG. 9), a threshold value is a distribution of element values having one maximum point and minimum point (ie, a distribution corresponding to one halftone dot). Although described as the matrix 81, of course, a three-dimensional LUT is generated based on a threshold matrix that is a distribution of element values having a plurality of maximum points and minimum points (that is, a distribution corresponding to a plurality of halftone dots). May be.

  In the above embodiment, an original image obtained by increasing the resolution of the original image and converting it to the proof resolution is acquired. However, when the resolution of the original image is higher than the resolution for the proof, in step S13, the original image is obtained. When the resolution is reduced and the resolution of the original image is the same as the resolution for proofing, the process of step S13 is omitted.

  The gradation level that can be taken by the pixels of the original image converted to the proof resolution does not necessarily match the gradation level that can be taken by the original image before conversion, and the number of gradations is increased in step S13 (or The original image may be reduced in resolution. Even in this case, in step S21, a plurality of binary flat mesh images each corresponding to the gradation levels that can be taken by the pixels of the original image after conversion are generated.

  The three-dimensional LUT generation unit 21 does not necessarily need to generate a plurality of binary flat mesh images respectively corresponding to all the gradation levels that can be taken by the pixels of the original image converted to the proof resolution. A multi-tone flat halftone image having a part of the tone level of the three-dimensional LUT may be thinned out within a range where the above problem does not occur.

  In addition, an image including a plurality of multi-tone flat halftone images is generated for each gradation level by arranging a plurality of generated binary flat halftone images in at least one of the row direction and the column direction. The proof image may be generated using a three-dimensional LUT based on these images.

  The threshold value matrix may be generated according to an original image according to a predetermined function, and further for each color plate. The unit area set in the original image is appropriately changed according to the generated threshold matrix.

  The computer 11 may display a proof image on the display 105 as necessary, and the displayed image may be handled as a print proof.

It is a figure for demonstrating the outline | summary of the process which produces the printed matter and print proof of a halftone image. It is a figure for demonstrating a mode that the high resolution original image is halftone-ized. It is a figure which shows the structure of a printing proof production system. It is a block diagram which shows the function structure of a proof image generation apparatus. It is a figure which shows the flow of the process which produces | generates a proof image. It is a figure which shows an example of the multi-tone flat halftone image of a three-dimensional LUT. It is a figure which shows the unit area for proofing of an original image. It is a figure which shows the flow of the process which produces | generates a three-dimensional LUT. It is a figure which shows a threshold value matrix. It is a figure which shows the binary flat-net image regarding the gradation level 128. FIG. It is a figure which shows the binary flat-net image regarding the gradation level 0. FIG. FIG. 6 is a diagram illustrating a binary flat screen image relating to a gradation level 32; FIG. 6 is a diagram illustrating a binary flat screen image relating to a gradation level 223; FIG. 6 is a diagram illustrating a binary flat screen image relating to a gradation level 225; It is a figure which shows a binary flat net image. It is a figure which shows the multi-tone flat halftone image. It is a figure which shows the flow of the process which produces | generates a three-dimensional LUT. It is a figure which shows the enlarged flat net image. It is a figure which shows a new flat net image. It is a figure which shows the multi-tone flat halftone image. It is a figure for demonstrating the relationship between a multi-tone flat halftone image and the unit area for proofs. It is a figure for demonstrating the other example of the process which produces | generates the multitone flat net image.

Explanation of symbols

7 Original image data 11 Computer 12 Printer 21 Three-dimensional LUT generation unit 22 Proof image generation unit 23 Proof image correction unit 71, 72 Original image 81 Threshold matrix 82, 82a to 82e, 83, 820, 820a, 820b, 830 Flat mesh image 84 Multi-tone flat halftone image 92 Program 104 Fixed disk 711 Halftone dot unit area 712, 721 Proof unit area S11, S12, S14 Steps

Claims (6)

A proof image generation device for generating a proof image of a halftone image representing a multi-tone original image,
A storage unit for storing original image data, and a threshold matrix used when halftone dots are generated for each predetermined unit area of the original image with high resolution;
For each of the gradation levels that can be taken by the pixels of the original image converted to the proof resolution, a binary flat image of the unit region having a uniform gradation level is generated using the threshold value matrix, A three-dimensional table having two-dimensional coordinates and gradation levels as parameters by further reducing the resolution of the binary flat mesh image in accordance with the proof resolution and further generating a multi-tone flat mesh image. Means for generating
In each unit area of the original image converted to the proof resolution, a multi-gradation proof image is generated by referring to the three-dimensional table based on the two-dimensional coordinates and gradation levels of each pixel. Means,
A proof image generating apparatus comprising: The proof image generating device according to claim 1,
The binary flat image is a row and a column (a is an integer of 2 or more), and the binary flat image is (n1 / m1) times (n1, m1 are natural numbers that are relatively prime) in the row direction. And, when the resolution is reduced to n1 <m1) and (n2 / m2) times (n2, m2 are relatively prime natural numbers and n2 <m2) in the column direction,
Means for generating the three-dimensional table;
The binary flat image is enlarged by n1 times in the row direction and n2 times in the column direction, and the greatest common divisor b1 between a and m1 and the greatest common divisor b2 between a and m2 are respectively obtained. A new flat net image is generated by arranging and calculating (m1 / b1) enlarged flat net images in the row direction and (m2 / b2) in the column direction, and m1 rows are generated from the new flat net image. A proof image generating apparatus that generates the multi-grayscale flat mesh image having a representative value of element values of m2 columns as an element value. A proof image generating device according to claim 1 or 2,
A proof image generating apparatus, wherein high-frequency components are removed before the resolution of the binary flat mesh image is reduced. A proof image generating device according to any one of claims 1 to 3,
A proof image generation apparatus, further comprising: a correction unit configured to perform correction depending on a proof printing apparatus on the proof image generated by the proof image generation unit. A proof image generation method for generating a proof image of a halftone image representing a multi-tone original image,
Preparing a threshold value matrix to be used when the original image data and the high-resolution original image are halftoned for each predetermined unit area;
For each of the gradation levels that can be taken by the pixels of the original image converted to the proof resolution, a binary flat image of the unit region having a uniform gradation level is generated using the threshold value matrix, A three-dimensional table having two-dimensional coordinates and gradation levels as parameters by further reducing the resolution of the binary flat mesh image in accordance with the proof resolution and further generating a multi-tone flat mesh image. Generating
In each unit area of the original image converted to the proof resolution, a multi-gradation proof image is generated by referring to the three-dimensional table based on the two-dimensional coordinates and gradation levels of each pixel. Process,
A proof image generation method comprising: A program for causing a computer to generate a proof image of a halftone dot image representing a multi-tone original image, the computer executing the program,
Preparing a threshold value matrix to be used when the original image data and the high-resolution original image are halftoned for each predetermined unit area;
For each of the gradation levels that can be taken by the pixels of the original image converted to the proof resolution, a binary flat image of the unit region having a uniform gradation level is generated using the threshold value matrix, A three-dimensional table having two-dimensional coordinates and gradation levels as parameters by further reducing the resolution of the binary flat mesh image in accordance with the proof resolution and further generating a multi-tone flat mesh image. Generating
In each unit area of the original image converted to the proof resolution, a multi-gradation proof image is generated by referring to the three-dimensional table based on the two-dimensional coordinates and gradation levels of each pixel. Process,
A program characterized by having executed.

Images