JP5469973B2 - Image conversion apparatus and image conversion program - Google Patents

Description

  The present invention relates to an image conversion apparatus and an image conversion program that determine the type of screening and the number of lines used in digital image photography, graphics, and the like, and convert an image according to the determination result.

  FIG. 1 is a diagram illustrating the concept of print images and dot gain for AM screening and FM screening. Conventionally, in a printing method that expresses an image in two states, such as offset printing, whether ink is applied or not, there are many methods for expressing the gradation of a multi-valued image such as a color print or a gray scale print. The value image is converted into a halftone dot image that is a collection of minute binary points (dots). This method is called screening, and represents the gradation of a multi-valued image by changing the way of dot arrangement (dot size, number, density). The halftone image is printed on a printing plate (hereinafter referred to as a printing plate), and the halftone image is printed on paper or the like using the printing plate. As a result, the printed matter appears as a multi-valued image when viewed macroscopically.

  There are various types of screening, but typical ones include AM screening and FM screening. The AM screen shown in FIG. 1A is a method of expressing light and shade by changing the size of dots arranged at equal intervals. On the other hand, the FM screen shown in FIG. 1B is a method of expressing light and shade by changing the density of dots having the same size. In AM screening and FM screening, the interval, size, and shape of dots can be arbitrarily set according to the image to be printed. For example, a circle, square, rhombus, or the like can be set as the dot shape. Conventionally, a method called hybrid screening in which different screening methods are mixed has been devised (see Patent Document 1).

  When performing offset printing, etc., as shown in FIGS. 1 (C) and 1 (D), the ink spreads around the dots due to the printing pressure, or the dots appear to be large due to irregular reflection inside the paper. Sometimes. These are called dot gains. At this time, the halftone dot of the printed material appears larger than the halftone dot size of the printing plate, and a phenomenon occurs in which the density at the time of observation becomes higher than the density obtained from the halftone dot.

  In actual printing, in order to avoid the effects of such a phenomenon, the dot gain characteristics of the printing press and printing materials (ink, paper) are obtained in advance by a printing test or the like, and the multivalued image is converted into a halftone image. When converting to, each dot of the halftone dot is configured to be smaller than the actual halftone dot area ratio. The dot gain changes depending on the type of screening and the interval and size of the dots. Therefore, when creating an image for a printing plate, the dot area ratio is corrected in consideration of the screening characteristics and the printing machine characteristics. . Then, a printing plate is created using the screened binary image (binary digital image), and printing is performed using the plate.

  At the time of printing, the operator monitors the color tone of the printed matter with reference to the reference printed matter printed by the calibration printer or the image (reference image) of the reference printed matter displayed on the monitor. The printing press is provided with a monitoring / inspection device such as a color tone monitoring device, a paper surface inspection device, and a printing plate misloading inspection device, and these devices also monitor the printed matter. When these screened binary images are input, these devices convert the binary image into a multi-value image equivalent to a target printed matter read by a sensor, and monitor the multi-value image. Used as a reference image for For this conversion, conversion called profile conversion is used. By converting a binary image using a profile in which color reproduction characteristics unique to each device are absorbed, an output result independent of the device can be obtained.

  As an apparatus for performing such processing, for example, there is an apparatus described in Patent Document 2. In this apparatus, a sensor device profile of the reference printing machine is created, and based on this, the mixed color halftone density for the pixel area of interest of each printed pattern is calculated from the plate-making data at each printing, and this is used for other printing. Set to a target color mixture halftone density for a printing machine (a printing machine to be controlled) (see Patent Document 2).

JP-A-2005-241810 JP 2006-239955 A

  However, the device profile used in the invention described in Patent Document 2 was obtained from the relationship between the original digital image and the sensor value obtained when a digital image was made under a certain condition and printed. Is. For this reason, when the characteristics during this period change and a different output value is obtained, there is a problem in that an accurate conversion value cannot be obtained and the print density is erroneously controlled. In particular, when the screening method is changed, even if digital images have the same area ratio, different printing results can be obtained, and this influence becomes great.

  Therefore, an object of the present invention is to provide an image conversion apparatus and an image conversion program capable of creating a highly accurate reference image even if the screening methods are different.

  The image conversion apparatus according to the present invention pays attention to the fact that the geometric properties of halftone dots constituting screening differ depending on the type of screening and the number of lines. The image conversion apparatus obtains the relationship between the feature amounts according to the difference in the geometrical properties of the halftone dots with respect to the screening of the binary digital image. Based on this relationship, the binary digital image is color space converted by the conversion means. By doing so, it is possible to perform an appropriate color space conversion according to the type of screening, the number of lines, etc., so that a highly accurate reference image can be created even if the screening method is different.

  In the present invention, characteristics such as halftone dot area ratio, boundary length of binary image forming halftone dots, halftone dot area, number of halftone dots, spatial frequency characteristics, etc., differ when the type of screening and the number of lines are different. It pays attention to. For each predetermined area of the binary digital image, the calculation means calculates the relationship between at least two feature values relating to the characteristics of the halftone dots. By using this feature value, the type of screening and the number of lines can be reliably identified. Thus, color space conversion corresponding to the screening can be performed. Further, even when a plurality of screenings are mixed on one sheet, it is possible to determine the difference and perform color space conversion according to the screening. Therefore, by using the image converted by the image conversion apparatus of the present invention, it is possible to accurately monitor the print finished state such as color tone monitoring, defect inspection, and printing error inspection.

  In the present invention, the binary digital image is divided into a plurality of predetermined areas by the dividing means. As a plurality of predetermined regions by the dividing means, for example, by dividing the reference image into one pixel unit, the calculating means allows the relationship between the feature amounts according to the difference in the geometric properties of the halftone dots to be obtained in one pixel unit. Thus, the conversion unit can appropriately perform color space conversion based on the relationship between the feature values. When the binary digital image is divided into units of one pixel of the reference image, if the color space conversion cannot be performed appropriately, the dividing unit is configured to divide the predetermined region into a region larger than the pixel size of the reference image. By setting to, color space conversion can be appropriately performed while suppressing local variations in pixel units.

  In the present invention, the type of screening is determined based on the determination function for determining the type of screening and the relationship between the feature quantities obtained by the calculation means, and the profile is selected according to the determination result. Thereby, the profile according to the kind of screening can be selected exactly, and conversion accuracy can be raised. A profile is a table for converting colors, which is used to match color characteristics between devices.

  In this invention, the binary digital image is divided into a plurality of image blocks, each image block is further divided into a plurality of predetermined regions, the type of screening is determined for each region, and the screening is performed based on the total result. The type is finally determined. The same type of screening is used for each image area, and even if there are misjudgments in several divided areas, the determination is based on the total result, so the type of screening can be accurately determined for each image area .

  In the present invention, when color printing is performed using a plurality of printing plates, a plurality of divided areas are determined for each plate, and the determination result of each plate is combined to finally determine the type of screening. Judgment. In this way, the screening type can be accurately determined even in the case of color printing. Further, when the type of screening is different for only one version data, it is possible to detect an error in the version data or to make a determination error such as an erroneous determination of the version.

  In the present invention, the calculating means obtains the first relationship between the feature amounts representing the difference in the geometric properties of the halftone dots for the image used for creating the profile, and for plate making used in actual printing. For the binary image, a second relationship between the feature amounts representing the difference in the geometric properties of the halftone dots is obtained. Further, the correction means calculates a correction value from the first relationship and the second relationship. The image conversion means reads the profile from the storage means and performs color space conversion of the binary digital image based on the profile. Then, the correcting unit corrects the image that has been subjected to the color space conversion by the image converting unit, based on the correction value. Therefore, since the color space conversion of the image can be performed according to the halftone dots, appropriate conversion can be performed. Further, when only one profile is stored in the storage means, a plurality of profiles are not required, so that it is possible to reduce time and labor for profile creation. Further, when a plurality of profiles are stored in the storage means, it is only necessary to correct the profile selected according to the feature amount, so that the error can be reduced.

  In the present invention, the color space is based on the relationship between the feature amounts calculated by the calculating means, the density information of the solid portion and the overlapped solid portion, the dot gain correction information, and the Neugebauer equation or the extended Neugebauer equation. Perform conversion. By doing so, it is not necessary to create a profile in advance, and it is possible to eliminate the labor and man-hour for creating the profile.

  In the present invention, based on the relationship between the feature amounts calculated by the calculating means, the halftone dot operating means performs color space conversion of the binary digital image and outputs the processing result to the outside. Thereby, when outputting the binary digital image which performed the color space conversion which processes a halftone dot to the apparatus which receives a binary image, the degradation of the image by image conversion, etc. can be prevented.

  According to the present invention, since appropriate color space conversion according to the type of screening can be performed, even when a plate using various screening methods is used, an accurate print result is predicted and a high-precision reference image is obtained. And print monitoring and proofreading can be performed using this reference image.

It is a figure which shows the concept of the printing image of AM screening and FM screening, and a dot gain. 1 is a block diagram illustrating a schematic configuration of a printing system. It is a conceptual diagram showing the method to extract or discriminate | determine the characteristic of a binary digital image in the image converter which concerns on embodiment of this invention. It is a figure for demonstrating the method of determining a several screening system. It is a graph which shows the relationship between the area of a halftone dot, and boundary length in several different screening. 1 is a block diagram illustrating a schematic configuration of an image conversion apparatus according to a first embodiment of the present invention. It is a flowchart for demonstrating operation | movement of the image conversion apparatus and color tone monitoring apparatus which concern on 1st Embodiment. It is a block diagram which shows the schematic structure of the image conversion apparatus which concerns on 2nd Embodiment of this invention. It is a flowchart for demonstrating operation | movement of the image converter which concerns on 2nd Embodiment of this invention. It is a block diagram which shows the schematic structure of the image conversion apparatus which concerns on 3rd Embodiment of this invention. It is a flowchart for demonstrating operation | movement of the image conversion apparatus which concerns on 3rd Embodiment of this invention. It is a graph which shows the relationship before the correction | amendment of the curve at the time of multi-valued (descreening), and after correction | amendment. It is a block diagram which shows the structure of the image conversion apparatus in the case of connecting a proofreading apparatus.

  FIG. 2 is a block diagram showing a schematic configuration of the printing system. A printing system 1 shown in FIG. 2 is a system that manages the color tone of a printed material. In the present invention, an image conversion device is provided in the printing system, and each image block (individual photograph or graphics) of an image area (an area including an image using a halftone dot (screening) such as a photograph or graphics). A reference image is generated according to the screening used. Then, using these reference images, the color tone is monitored.

  The printing system 1 includes an image server 13 that stores and provides image files, a CTP (Computer to Plate) 17 that produces a printing plate, and an offset rotary press (hereinafter simply referred to as a rotary press) 19 that prints printed matter. ing. Further, the printing system 1 determines the type of screening used in the image area, and outputs an image converted according to the result to a monitoring device such as the calibration device 20K or the color tone monitoring device 21, A calibration device 20K that outputs a calibration image created from an image output from the image conversion device 20, a color tone monitoring device 21 that monitors the color tone of the printed matter, and a monitor 22 that displays the ink state estimated by the color tone monitoring device 21 are provided. ing. Further, the printing system 1 includes an ink key control device 23 that controls the opening degree of the ink keys 72 </ b> A to 72 </ b> H, and optical sensors 25 </ b> F and 25 </ b> R that capture images of printed matter printed by the rotary press 19.

  Features and operations of each part of the printing system 1 are as follows.

  The image server 13 accumulates image data for printing (image data in 1 bit TIFF format as an example).

  The CTP 17 reads image data for printing (binary halftone image data [image data in 1-bit TIFF format]) from the image server 13, and produces a printing plate for each ink of the rotary press 19.

  The rotary press 19 is set with a printing plate manufactured by the CTP 17 and supplies a web-like continuous paper P from the roll paper R. The printing unit 73 is black (hereinafter referred to as K) / cyan ( Cyan: hereinafter referred to as C.) Magenta (hereinafter referred to as M) Yellow (Yellow: hereinafter referred to as Y) ink is supplied from a plurality of ink keys 72A to 72H, and the continuous paper P Print letters and designs on both sides. The ink supply order is not limited to this order, and the ink is supplied in a different order depending on the newspaper company or the printing factory.

  In the rear stage (downstream side) of the printing unit 73, an optical sensor 25F is provided to face the front side of the continuous paper P, and an optical sensor 25R is provided to face the back side of the continuous paper P. The optical sensors 25F and 25R take images of the front surface (front cylinder side) and the back surface (back cylinder side) of the running printing paper surface that has been printed with CMYK or other color inks by the printing unit 73, so that a plurality of wavelengths can be obtained. The image data (sensor value) of the area is output for each pixel.

  As an example, the optical sensors 25F and 25R receive light irradiated from a light source (not shown) and reflected from the surface of the continuous paper P, and receive red (hereinafter referred to as R) and green (hereinafter referred to as R). G.) ・ Blue (Blue: hereinafter referred to as “B”) ・ Infrared rays (hereinafter referred to as “Ir”). , RGBIr may be indicated), and image data (sensor values) in four wavelength regions is output for each pixel. The optical sensors 25F and 25R output the reflected light amount value as digital data as an example. The light source (not shown) may irradiate light including all wavelengths of visible light, may irradiate light in only four wavelength regions of R, G, B, and Ir, or may be a single light source. A plurality of light sources may be used. When printing is performed using inks of other colors in addition to CMYK, the printed pattern may be imaged in the wavelength region of RGBIr or in another wavelength region.

  Further, as described above, the optical sensors 25F and 25R capture images in the four wavelength regions of RGBIr and output sensor values (actually measured image data) in each wavelength region for each pixel. The sensor values in the wavelength region are collectively referred to as RGBIr values.

  The image conversion apparatus 20 determines the type of screening used for the image area in the binary image data (1 bit TIFF) acquired from the image server 13. Then, based on the determination result, an image close to the print finished state is synthesized and output to the color tone monitoring device 21. As described above, an area including an image using a halftone dot (screening) such as a photograph or graphics is referred to as an image area. An area for displaying an image of a character that does not use halftone dots is referred to as a character area. In the present application, although description is omitted, color tone management is performed in a character area by a known method.

  An ink key control device 23 is connected to the rotary press 19.

  The calibration device 20K processes the image input from the image conversion device 20 and performs a calibration process for the print image.

  The color tone monitoring device 21 monitors the newspaper image printed by the rotary press 19 with the optical sensors 25F and 25R, and outputs a control signal to the ink key control device 23. A monitor 22 is connected to the color tone monitoring device 21.

  The ink key control device 23 adjusts the opening of each of the ink keys 72A to 72H according to a control signal from the color tone monitoring device 21 or according to an operation performed by the operator according to the display content of the monitor 22.

  In the rotary press 19, characters, photographs, and graphics (pictures) are printed on both sides of the printed matter, and these are imaged by the optical sensors 25F and 25R to adjust the color tone of both sides of the printed matter. In the following description, in order to simplify the description, only the description of performing color tone monitoring / management by imaging the surface of the printed matter with the optical sensor 25F will be described.

  Next, an outline of processing performed by the image conversion apparatus that is a characteristic configuration of the present invention will be described. The image conversion apparatus 20 converts an image by any one of the following three methods. In the following description, a plurality of dots are referred to as halftone dots not only for AM screening but also for FM screening.

  (1) The image conversion apparatus 20 receives binary digital image data for printing such as a 1-bit TIFF image, and has geometric properties (halftone dot area ratio, halftone dot area, halftone dot) for halftone dots in the image. Point boundary length, number of halftone dots per unit area, spatial frequency characteristics, etc.), and two or more different feature quantities representing a difference in the geometric properties of halftone dots are analyzed. Seeking a relationship between. Then, an appropriate profile is selected based on the relationship between the feature amounts for the image after the binary digital image is multivalued (descreened), and color space conversion is performed using the selected profile.

  (2) The image conversion apparatus 20 obtains a relationship between feature amounts representing a difference in geometric properties of the binary digital image used for creating the reference profile. In addition, regarding a binary digital image for plate making used in actual printing, a relationship between feature amounts representing a difference in the geometric properties of halftone dots is obtained. Then, a correction value is calculated from the two relationships. Further, color space conversion is performed on the image obtained by converting the binary digital image into multi-values using a reference profile, and the image subjected to color space conversion using the correction value is corrected.

  (3) The image conversion apparatus 20 does not prepare a profile, and applies a function or LUT that represents a relationship between, for example, a dot area ratio, a boundary length, a dot gain, or an effective area ratio to a binary digital image, An effective area ratio taking dot gain into consideration is obtained, and a well-known Neugebauer equation is applied to this effective area ratio. Then, the RGBIr reference image is synthesized by multiplying this by the reflectance of each spectrum of the solid multiplication of four colors (solid portion and overlapping solid portion 16 colors).

  The image conversion apparatus 20 performs any of the processes described above, and from the binary digital image (CMYK image), an image (RGBIr) that is close to the image of the print finished state required by the print state monitoring device or the like captured by the sensor. A reference image). In the printing system 1, the color tone monitoring device 21 and other monitoring devices (not shown) (for example, a defect inspection device, a printing error inspection device, etc.) finish printing using the image synthesized by the image conversion device 20. Monitor status. As will be described in detail later, the image conversion device 20 is configured to perform profile selection and image correction before descreening, and output this image to the calibration device 20K and other rotary press (not shown). It is also possible.

  A profile is a table for converting colors, which is used to match color characteristics between devices.

  Hereinafter, three embodiments of the image conversion apparatus will be described in order.

  Here, in the following description, in order to simplify the description, it is assumed that square dots (squares) are used as halftone dots. Square dots are connected with halftone dots at an intermediate density of 50%, and the shape is a checkered pattern.

  Of course, it is possible to use chain dots (diamonds), round dots (circular), or other shapes as halftone dots. In this case, since the density at which the dots are connected is different from that of the square dots, it is preferable to obtain a relationship according to the characteristics of the dots.

[First Embodiment]
In the following, first, characteristic processing performed by the image conversion apparatus according to the first embodiment of the present invention will be described, and then specific configuration and processing of the image conversion apparatus will be described. In the following description, the image conversion apparatus 20 according to the first embodiment is referred to as an image conversion apparatus 20A. FIG. 3 is a conceptual diagram showing a method for extracting or discriminating the characteristics of a binary digital image (plate making image) in the image conversion apparatus according to the embodiment of the present invention.

In general, as elements representing features of the shape of halftone dots (features representing differences in geometric properties), area, number of halftone dots, area ratio (number of halftone dots × halftone dot area / unit region) Area), boundary length of binary image forming halftone dots, inclination (angle) of halftone dots, circularity (4π · area / (boundary length) ² ), complexity ((boundary length) ² / area), The moment and the center of gravity (the principal axis of inertia indicating the position of the center of gravity and the direction of the figure), the ferret diameter (the length of the rectangle circumscribing the figure, the length of the horizontal side), and the like are known.

  In this embodiment, at least two of the above feature quantities representing the features of the halftone dot portion of the binary digital image are extracted, and the relationship between these feature quantities is obtained. For example, at least two of the halftone dot area ratio for each predetermined region, the boundary length of the binary image forming the halftone dot, the halftone dot inclination (angle), the halftone dot area, and the number of halftone dots are extracted. Thus, the relationship between these feature quantities is obtained. The relationship between the feature amounts may be obtained by experiments or the like. In addition, when the relationship between a plurality of feature quantities can be expressed as an expression (function), the relational expression may be calculated. Then, an appropriate profile is selected based on the relationship between the feature amounts, and color space conversion is performed based on the selected profile.

  Details of each feature amount are as follows.

  The predetermined area is, for example, a 6 mm square unit as one area (unit area). If the sensor resolution is 1.2 mm square, this corresponds to a sensor resolution of 5 × 5 squares.

  The halftone dot area ratio represents the ratio of the area of all halftone dots per unit region (unit area).

  The boundary length of a binary image forming a halftone dot (hereinafter referred to as the halftone dot boundary length) is the length of the boundary line between a dot and a blank portion in a unit area. For example, when dots are not in contact, this corresponds to the total perimeter of each dot, and when dots are in contact, this corresponds to the sum of the perimeter of each blank portion. When obtaining the boundary length of a halftone dot, the same calculation processing is performed by edge detection or the like in image processing regardless of the area ratio. That is, it can be calculated by a method of extracting an edge by edge analysis and measuring a dot length of a portion connected in a bag shape. Moreover, it is computable also by a well-known method, such as the method of Unexamined-Japanese-Patent No. 2004-287930.

  The gradient (angle) of the halftone dot affects the calculation of the boundary length of the halftone dot. For example, when the halftone dot is a square, the halftone dot is a collection of small dots (pixels), and shaggy around the halftone dot is conspicuous. If the area is calculated by adding the boundary lengths of the shaggy parts, it will be larger than the actual area, so this effect is taken into account. In addition, since the dot gain changes according to the type of screening and the linear density, this feature is a value related to dot gain correction.

  When the boundary length is used as the feature amount, the boundary length may be corrected according to the angle of the halftone dot as shown in FIG. For example, when the tilt angle is 45 degrees, the actual inter-lattice distance of the dots becomes √2 times, and accordingly, it is corrected to √2 times accordingly. If the tilt angle is 60 degrees, √ (3/2 ) It should be corrected to double.

  By representing the number of halftone dots (dots) for each unit area, the fineness of the plate-making image can be shown. Regardless of the fineness of the halftone dots, in order to represent the same density, the area of the halftone dots (dots) per unit area must be equal. If the size of the halftone dots (dots) is fine, the halftone dots Need to increase the number of. In this case, if the fine halftone dot (dot) is increased, the boundary length of the halftone dot is increased, so that the dot gain is changed.

  When a halftone dot area ratio and a boundary length are selected as feature quantities, the relationship between them is as follows.

  If the halftone dot area ratio is the same area ratio, it represents the same density. In general, if the area ratio is the same, FM screening often includes more dots (fine dots) than AM screening. In this case, the sum of the boundary lengths (halftone dot boundary lengths) of each dot of the halftone dot is longer in the FM screening. Therefore, discrimination between AM screening and FM screening, and discrimination of screening fineness can be characterized by using the relationship between the halftone dot area ratio A per unit region and the halftone dot boundary length B. In the following, for AM screening, the relationship is modeled using a flat knitting dot as an example. Note that, in a certain dot area ratio, each dot in the unit region has the same size.

In the case of a low density area: In the case of square dots, if the area ratio of halftone dots is less than 50%, each dot of the halftone dots will not touch. In this case, the dot area ratio A and the boundary length B are √A∝B. Each dot of a halftone dot is a square, where L is one side of the dot, N is the number of dots, and S is the length of one side of the unit area.
Dot area ratio A = (N · L ² ) / S ² ,
One side of the dot L = (√A) · S / (√N),
Halftone dot boundary length B = N · 4L,
B = N · 4 · (√A) · S / (√N) = 4 · (√N) · S · √A
If K1 = 4 · (√N) · S,
B = K1 · √A.
“·” Is multiplication, and “(√x)” is the square root of x (the same applies hereinafter).

In the case of a high density area: In the case of square dots, when the area ratio of halftone dots is 50% or more, each dot of the halftone dots contacts (connects). The area ratio 1-A and the boundary length B of the blank portion are √ (1-A) ∝B. The blank part is a square, and if one side is L and the number of blank parts is M,
Area ratio 1-A = (M · L ² ) / S ^{2 of} the blank portion
Dot area ratio A = 1- (M · L ² ) / S ² ,
One side L = {√ (1-A)} · S / (√M)
Blank part boundary length B = M · 4L,
B = M · 4 · {√ (1-A)} · S / (√M)
= 4 · (√M) · S · {√ (1-A)}
If K2 = 4 · (√M) · S,
B = K2 · √ (1-A).

In summary,
Boundary length of halftone dot B = K1 · √A (0% ≦ A <50%)
However, (K1 = 4 · (√N) · S)
Boundary length of blank part B = K2 · √ (1-A) (50% ≦ A ≦ 100%)
However, (K2 = 4 · (√M) · S)
It becomes.

  A function representing this relationship is referred to as function F1, and B = F1 (A).

  On the other hand, for FM screening, a function is obtained based on a curve obtained for the relationship between the area ratio of halftone dots and the boundary length. For example, for FM screening, in a region where the area ratio until the halftone dots overlap is low, the halftone dot area ratio is proportional to the number of dots, and the boundary length B is also proportional to the number of dots. Therefore, the boundary length B increases in proportion to the dot area ratio A. In the region where the area ratio where the halftone dots overlap is high, the boundary length B becomes smaller in proportion to [1-dot area ratio (blank area ratio) A]. A function representing this relationship is referred to as function F2, and B = F2 (A).

  Since the relationship between the dot area ratio and the boundary length actually depends on the screening method, a function is obtained based on an experimentally obtained curve.

  If the size of the halftone dots is small, it is necessary to increase the number of dots in inverse proportion to the same area ratio. When the number of dots increases, the boundary length of the dots increases accordingly. Therefore, it can be said that the proportionality constant K and the function (curve) F (A) specified thereby are features representing the fineness of the halftone dots.

  When the functions F1 and F2 representing the relationship between the area ratio and the boundary length are obtained for the AM screening and the FM screening, a function of an intermediate value of both functions is set as a threshold function.

  For example, as shown in FIG. 3B, an intermediate value function {F1 (A) + F2 (A)} / 2 between the functions F1 (A) and F2 (A) is set as a threshold value. Thereby, the kind of screening can be easily determined by comparing the value of the boundary length with the threshold value. As described above, this threshold function may be an arithmetic mean, but may be a geometric mean, or any function can be used as long as it is a function between both similar functions.

  Accordingly, whether the AM screening or the FM screening is determined is determined by using the intermediate value function {F1 (A) + F2 (A)} / 2 as a threshold curve (threshold function for screening determination) and above. FM screening is determined, and if lower than that, AM screening is determined. This threshold curve can be easily obtained from AM screening and FM screening used in a factory, for example.

  In the example shown in FIG. 3B, FM screening is performed when the boundary length of the halftone dot is larger than the threshold function 641 (B = {F1 (A) + F2 (A)} / 2) regardless of the area ratio of the halftone dot. Further, if the boundary length of the halftone dot is smaller than the threshold function 641, the AM screening is performed.

  In the present invention, a digital image is handled. In the above description, an equation for an analog image is shown so that it can be easily understood. The above formula is actually conceptual and is not suitable for applying to digital images. Therefore, in practice, a digital image function for obtaining the boundary length of halftone dots is obtained. When an expression (function) representing the relationship between the dot area ratio and the boundary length cannot be obtained, an actual measurement value is used.

  Of course, it is possible to select another combination as the feature amount representing the feature of the halftone dot portion of the binary digital image. Each feature amount may be converted from another feature amount. For example, if the dot area and the number of dots in a predetermined area are determined, the dot area ratio of the predetermined area is determined. Therefore, the dot area ratio of the predetermined area is converted into the dot area and the number of dots in the predetermined area. Can do.

  Next, in the above description, two screening determination methods have been described, but a plurality of screening determination methods will be further described. 4A and 4B are diagrams for explaining a method of determining a plurality of screening methods. FIG. 4A shows a method of determination using a threshold value of a curve, and FIG. 4B shows a plurality of screening methods. Shows an example of a print image including.

  When discriminating a plurality of known screenings, a function representing the relationship between the area ratio and the boundary length is obtained in advance for known screenings. Then, an intermediate value function between adjacent functions is set as a threshold function. Thereby, the kind of screening can be easily determined by comparing the value of the boundary length with the threshold value.

  For example, as shown in FIG. 4A, in the case of discriminating three screenings, a threshold value may be set as follows. First, functions F3 (A), F4 (A), and F5 (A) (curve shown in FIG. 4A) representing the relationship between the area ratio of known screening and the boundary length are obtained. Subsequently, an intermediate value function {F3 (A) + F4 (A)} / 2 between adjacent functions F3 (A) and F4 (A) is set as a threshold value. Further, a curve of an intermediate value function {F4 (A) + F5 (A)} / 2 between adjacent functions F4 (A) and F5 (A) is set as a threshold value. Thereby, the kind of screening can be easily determined by comparing the value of the boundary length with the threshold value. As in the case described with reference to FIG. 3, these functions may be arithmetic averages, but may be geometric averages, and any function may be used as long as the functions are similar to each other. It is also possible.

  For example, as shown in FIG. 4B, the binary image data includes a character area (block 1001 and block 1006) and an image area (image blocks 1002, 1003, 1004, and 1005). , 1004 is the screening S1 (function F3 (A)), the image block 1005 is the screening S2 (function F4 (A)), and the image block 1002 is the screening S3 (function F5 (A)), as described above. By setting, the type of screening used for each image block can be determined.

  Accordingly, by using profile selection and image data correction, which will be described later, it is possible to prevent a color tone from being different only by a part of the print image (for example, part 1007 of the image) as in the past. .

  Next, a case where the type of screening is determined using an actual measurement value calculated from the digital data of the plate will be described.

FIG. 5 is a graph showing the relationship between the area of halftone dots and the boundary length in a plurality of different screenings. FIG. 5 plots the relationship between halftone dot area and boundary length for a plurality of halftone dot test charts formed by AM screening or FM screening. However, in the graph shown in FIG. 5, the horizontal axis uses not the area ratio but the size of 100 dots (Pixel) angle itself, and 100 ² = 10000 corresponds to the area ratio of 100%. The boundary length on the vertical axis represents the total number of consecutive dots (Pixel).

  Note that if a plurality of screenings are further included in one plate-making image, it may occur that each screening cannot be identified. In this case, in order to extract the characteristics of the plate-making image, it is not always necessary to specify the estimated value of the proportionality constant K from the actual value curve as shown in FIG. Even when the proportionality constants K1 and K2 described with reference to FIG. 3 are not obtained, dot gain correction is performed only with the data set (A (halftone dot area ratio), B (boundary length)) actually measured as shown in FIG. Can be used. That is, since the screening is determined based on the relationship between the boundary length of halftone dots and the area, it is possible to determine the type of screening by using only the actual measurement value and perform correction.

  Since (A, B) is obtained by individually processing the image of the halftone dots actually arranged, the function F (A) can be obtained only as a discrete value function. In mounting on the image conversion apparatus 20, B = F (A) is stored in the storage unit 201 as a discrete value data set or a lookup table, and is used by interpolation during implementation.

  In the present invention, a relational expression between the halftone dot area (average area) A and the spatial frequency characteristic f may be calculated as the feature quantity. For example, a binary image is subjected to Fourier transform for each unit region, and a spatial frequency f that is a peak is obtained. Considering that halftone dots are repetitive patterns, the spatial frequency f is proportional to the number of halftone dots when considered in one dimension. Actually, since the image area is two-dimensional, the square of f is an amount proportional to the number of halftone dots. Further, the average value W of the light and shade of the binary image is obtained. The average value W may be a DC term obtained by Fourier transforming a binary image for each region as described above. The following relationship holds between the average area A of each halftone dot, the average value W of the density of the binary image, and the spatial frequency f.

A∝W / f ² (Formula 1)
Therefore, the type of screening can be determined based on (Equation 1).

  The above description is a case where the halftone dots do not overlap, and the above relationship does not hold when the halftone dots overlap. In addition, the above relationship does not hold when black and white is reversed (when it exceeds 50%). In such a case, it may be judged by comparing with a relational expression obtained by experiments or the like.

  Next, the configuration and operation of the image conversion apparatus according to the first embodiment of the present invention will be described. FIG. 6 is a block diagram showing a schematic configuration of the image conversion apparatus according to the first embodiment of the present invention. As shown in FIG. 6, the image conversion apparatus 20A includes a dividing unit (dividing unit) 200A, a calculating unit (calculating unit) 202A, a storage unit (storage unit) 201A, a determining unit (determining unit) 203A, and an image processing unit. 205A and a conversion unit (image conversion unit) 206A (the storage unit 201A to the conversion unit 206A correspond to the conversion unit).

  When the CMYK binary digital image data (1 bit TIFF) is input from the image server 13 outside the apparatus, the dividing unit 200A separates a plurality of image areas (image blocks) included in the binary digital image, and The region (image block) is further divided into a plurality of regions (predetermined regions). For example, the dividing unit 200A divides the image area into one pixel unit of the reference image according to the setting. It can also be set to divide into predetermined areas larger than the pixel size of the reference image.

  The storage unit 201A stores a threshold function (determination function) for determining the type of screening by the determination unit 203A and a plurality of profiles for performing color space conversion by the conversion unit 206A.

  When the CMYK binary digital image data (1 bit TIFF) divided into predetermined regions is input from the dividing unit 200A, the calculation unit 202A analyzes the halftone dots used in the image regions in the binary digital image. Thus, the relationship between the feature quantities representing the difference in geometric properties is obtained for each region. For example, when the calculation unit 202A is set in advance so as to obtain the relationship between the dot area ratio and the boundary length as the relationship between the feature amounts, the calculation unit 202A obtains the relationship between them.

  The determination unit 203A confirms the relationship between the feature amounts obtained by the calculation unit 202A, the threshold function read from the storage unit 201A, and the magnitude relationship, and determines each of the divided image regions (predetermined regions). Determine the type of screening. Then, the determination unit 203A aggregates the determination results, determines the most frequent screening type as the screening type of the image region, and outputs information on the screening type to the conversion unit 206A.

  When the CMYK binary digital image data (1 bit TIFF) divided into a plurality of regions (predetermined regions) is input from the dividing unit 200A, the image processing unit 205A descreens these binary digital images (multivalued). ). Then, the image processing unit 205A outputs the image subjected to the above processing to the conversion unit 206A. Descreening is a process of converting a binary image into a multi-value (multi-gradation) image.

  The conversion unit 206A selects and reads a profile from the storage unit 201A based on the screening type information output by the determination unit 203A. Then, the image for each predetermined area output by the image processing unit 205A using the read profile is subjected to profile conversion (color space conversion), and an RGBIr image is output.

  By configuring in this way, even if the type of screening is misjudged for a part of a plurality of regions (predetermined regions), the type of screening is determined for each photographic region because it is determined based on the total result. Accurate judgment can be made.

  Similarly, when hybrid screening is included in the type of screening used in the image region, the relationship between the feature amounts representing the difference in the geometrical properties of the halftone dots is obtained for each screening. In the same manner as described above, an appropriate profile is selected based on the relationship between these feature values, and color space conversion is performed based on the selected profile.

  When the binary digital image is composed of a plurality of plates, the determination unit 203A performs determination for each plate, and further determines the type of photographic area screening by combining the determination results of each plate. It is good to constitute so that. Thereby, even in the case of color printing, the type of screening can be accurately determined.

  Further, when the type of screening is determined as described above, the determination results for each of a plurality of regions may be simply summed, and the one having the largest total number may be determined as the type of screening. Alternatively, an evaluation function may be created in advance, and the weights assigned to values by this evaluation function may be summed to determine the type of screening. It should be noted that the evaluation function is preferably created based on the result of an experiment conducted in advance.

  In FIG. 6, only the color tone monitoring device 21 is shown as a device for monitoring the finished state of printing. However, the printing system 1 is a device for monitoring the finished state of printing, such as a defect inspection device or a printing error inspection device. The structure containing this may be sufficient.

  Next, operations of the image conversion apparatus and the color tone monitoring apparatus according to the first embodiment will be described based on flowcharts. FIG. 7 is a flowchart for explaining operations of the image conversion apparatus and the color tone monitoring apparatus according to the first embodiment.

  First, the image conversion apparatus 20A receives binary digital image data (CMYK 1-bit image, 1-bit TIFF) from the image server 13 (s1). Subsequently, the dividing unit 200A divides each image block of the image region into a plurality of regions (predetermined regions), and outputs them to the calculating unit 202A and the image processing unit 205A (s2). Further, the calculation unit 202A obtains the relationship between the area ratio of the halftone dots and the boundary length for each predetermined region, and outputs this to the determination unit 203A (s3). For example, the image conversion apparatus 20A obtains the relationship between the dot area ratio and the boundary length for each predetermined region, and outputs the relationship to the determination unit 203A.

  Further, when the relationship between the halftone dot area rate and the boundary length is input from the calculation unit 202A, the determination unit 203A receives a threshold function for determining the type of screening from the storage unit 201A (the halftone dot area rate and the boundary). A function or look-up table representing a long relationship is read (s4). Then, the determination unit 203A determines the type of screening in this region based on the magnitude relationship between this threshold function and the relational expression between the area ratio of the halftone dots and the boundary length input from the image server 13 ( s5). Note that by preparing a plurality of these functions, it is possible to determine the difference in the number of lines.

  As a specific example of the threshold function, area ratio and boundary length functions F1 (A) and F2 (A) are obtained for AM screening and FM screening, respectively, as described with reference to FIG. Then, an intermediate value function {F1 (A) + F2 (A)} / 2 between the functions F1 (A) and F2 (A) is set as a threshold function. AM screening is performed below this threshold function, and FM screening is performed above this threshold function. Further, the value of the proportionality constant K changes depending on the number of lines. The threshold function may be a mathematical expression obtained by calculation or an experimental expression obtained in advance.

  Based on the above determination, it is determined whether the area is AM screening or FM screening, and the number and shape of AM screening lines.

  On the other hand, the image processing unit 205A multivalues (descreens) the binary digital image (CMYK 1-bit image), and outputs the CMYK multivalued image to the conversion unit 206A (s6).

  The conversion unit 206A selects a profile corresponding to the determination result of the determination unit 203A from the storage unit 201A, converts the CMYK multivalued image output from the image processing unit 205A into an RGBIr reference image, and converts the profile to the color tone monitoring device 21. Output (s7).

  After printing is started, the color tone monitoring device 21 compares the RGBIr reference image output from the image conversion device 20 with the RGBIr image read by the optical sensor 25 (s11). When the difference in color tone between the two images is converging (s12: Y), the color tone monitoring device 21 continues to compare and check the images without adjusting the ink key. On the other hand, when the difference in color tone between the two images is increasing (s12: N), a control signal is output to the ink key control device 23 so that the color tone difference converges (the color tone becomes equal). , The opening of the ink key of CMYK is adjusted (s13). And the process after step s11 is performed.

[Second Embodiment]
Next, a characteristic process performed by the image conversion apparatus according to the second embodiment of the present invention will be described, and then a specific configuration and process of the image conversion apparatus will be described. In the following description, the image conversion apparatus 20 according to the second embodiment is referred to as an image conversion apparatus 20B.

  In this embodiment, with respect to a binary digital image (1 bit Tiff image) used to create a reference profile (one reference profile), the relationship between the feature quantities representing the difference in the geometric properties of halftone dots (first The relationship). Further, image data of a printing plate used for actual printing is acquired, and a relationship (second relationship) between the feature amounts is obtained for the image data. Then, a correction value is calculated from the first relationship and the second relationship. Then, the multi-value image obtained by converting the plate data into multi-values is subjected to profile conversion (color space conversion) using the reference profile, and the converted image is corrected using the correction value to generate a reference image.

  In the following description, as in the first embodiment, a case where the relationship between the area ratio of halftone dots and the boundary length is obtained as a feature amount will be described as an example.

  First, the image conversion apparatus 20B determines the relationship (function) (first relationship) between the halftone dot area ratio A1 and the boundary length B1 (A1) for each version of the test chart image used to create the reference profile. Ask. For example, ECI2002 or ISO12642 may be used as the test chart image. B1 (A1) represents a function of the dot area ratio and the boundary length (hereinafter the same).

  Further, the relationship (function) (second relationship) between the halftone dot area ratio A2 and the boundary length B2 (A2) is obtained for each plate of the image actually used for printing.

  Subsequently, the difference between the first relationship and the second relationship is obtained. For example, in the first relationship and the second relationship, the boundary length at the area ratio A2 is obtained, and the difference B2 (A2) −B1 (A2) is obtained for each color. This is defined as δBc, δBm, δBy, and δBk.

  The difference in dot gain (increase / decrease in area) can be expressed as a function of the difference in halftone dot boundary length and the area ratio. If this is assumed to be f (δB, A2), the increase / decrease in the area is expressed by f (δBc, A2), f (δBm, A2), f (δBy, A2), f (δBk, A2). It is assumed that this function is normalized to take a value of 1 when it occupies the entire area. Further, the function of the difference between the boundary lengths of the halftone dots and the area ratio may be obtained by experiments or the like.

  A correction value is obtained by multiplying the increment (decrement) difference by the area ratio obtained by the well-known Neugebauer equation or the extended Neugebauer equation, and the correction value is used for correction. For example, corresponding to C (indigo), for each area ratio of W, M, Y, K, MK, YM, YK, YMK obtained from each area ratio of CMYK (including dot gain correction at the time of reference profile) by Neugebauer formula F (δBc, A2), and for each area, the difference between the reflectance when there is no reflected light amount C for each color of RGBIr and the reflectance when there is C is obtained for each color obtained from the previous reference profile. The RGBIr value after profile conversion is corrected by subtracting it from the reflectance.

  Similarly, the amount of increase / decrease in the area ratio is corrected for M, Y, and K by the Neugebauer equation.

  The reflectivity of RGBIr obtained from the reference profile is Rr, Rg, Rb, Rir, and the reflectivity for RGBIr at each combination of CMYK (this is obtained in advance by experiment) is as follows. The reflectances are Rr, Rg, Rb, and Rir corresponding to R, G, B, and Ir, respectively. In addition, each of the combinations of CMYK colors w (white), c, m, y, k, cm, cy, ck, my, mk, yk, mcmy, cyk, cmyk The reflectance of RGBIr corresponding to the combination of CMYK is used.

  About R: Rrw, Rrc, Rrm, Rry, Rrk, Rrcm, Rrcy, Rrck, Rrmy, Rrmk, Rryk, Rrmy, Rrcmk, Rrcyk, Rrmyk, Rrmyk.

  About G: Rgw, Rgc, Rgm, Rgy, Rgk, Rgcm, Rgcy, Rgck, Rgmy, Rgmk, Rgyk, Rgcmy, Rgcmk, Rgmyk, Rgmyk, Rgcmyk.

  About B, Rbw, Rbc, Rbm, Rby, Rbk, Rbcm, Rbcy, Rbck, Rbmy, Rbmk, Rbyk, Rbcmy, Rbcmk, Rbmyk, Rbmyk, Rbcmyk.

  Ir is defined as Riw, Ric, Rim, Riy, Rik, Ricm, Ricy, Rick, Rimy, Rimk, Riyk, Ricmy, Ricmk, Ricyk, Rimyk, Rimyk.

  Further, each area ratio calculated from Neugebauer's equation is set as Aw, Ac, Am, Ay, Ak, Acm, Acy, Ack, Amy, Amk, Ayk, Acmy, Acmk, Acyk, Amyk, Amyk, and Amyk. The reflectance of RGBIr is corrected by calculating the following equation.

Rr '= Rr
− {(Rrw−Rrc) · Aw + (Rrm−Rrcm) · Am
+ (Rry−Rrcy) · Ay + (Rrk−Rrck) · Ak
+ (Rrmy−Rrcmy) · Amy + (Rrmk−Rrcmk) · Amk
+ (Rryk-Rrcyk) · Ayk
+ (Rrmyk-Rrcmyk) · Amyk} · f (δBc, A2)
− {(Rrw−Rrm) · Aw + (Rrc−Rrcm) · Ac
+ (Rry−Rrmy) · Ay + (Rrk−Rrmk) · Ak
+ (Rrcy−Rrcmy) · Acy + (Rrck−Rrcmk) · Ack
+ (Rryk-Rrmyk) · Ayk
+ (Rrcyk−Rrcmyk) · Acyk} · f (δBm, A2)
− {(Rrw−Rry) · Aw + (Rrm−Rrmy) · Am
+ (Rrc-Rrcy) .Ac + (Rrk-Rryk) .Ak
+ (Rrcm−Rrmmy) · Acm + (Rrmk−Rrmyk) · Amk
+ (Rrck-Rrcyk) · Ack
+ (Rrcmk−Rrcmyk) · Amck} · f (δBy, A2)
− {(Rrw−Rrk) · Aw + (Rrm−Rrmk) · Am
+ (Rry−Rryk) · Ay + (Rrc−Rrck) · Ac
+ (Rrmy−Rrmyk) · Amy + (Rrcm−Rrmyk) · Acm
+ (Rrcy-Rrcyk) · Acy
+ (Rrcmy−Rrcmyk) · Acmy} · f (δBk, A2)
Further, the correction of the reflectance of G, B, and Ir is also corrected by the same equation as above. (Replace with Rrxx → Rgxx above.)
Then, the multi-valued image obtained by multi-value / descreening the plate data is profile-converted by the reference profile, and the profile-converted image is corrected by the above correction formula and output as a reference image.

  In this correction, the difference in dot gain is not significantly large, and the correction is made within a range in which it is not always necessary to consider multiplication of a plurality of color increments. When dealing with a larger difference in dot gain, it is desirable to consider these multiplications. In the above description, the correction is performed using a linear formula, but other correction formulas or correction using a lookup table may be used.

  Next, the configuration and operation of the image conversion apparatus according to the second embodiment of the present invention will be described. FIG. 8 is a block diagram showing a schematic configuration of an image conversion apparatus according to the second embodiment of the present invention. As shown in FIG. 8, the image conversion apparatus 20B deletes the determination unit 203A of the image conversion apparatus 20A, newly provides a correction unit (correction unit) 207B after the conversion unit 206B, and the correction calculated by the calculation unit 202B. This is a configuration for correcting the image converted (output) by the conversion unit 206B based on the value.

  When the CMYK binary digital image data (1 bit TIFF) is input from the image server 13, the dividing unit 200 </ b> B separates a plurality of image areas (image blocks) included in the binary digital image, and further separates each image area. Divide into a plurality of regions (predetermined regions).

  When the binary digital image data (1 bit TIFF) for each predetermined region is input from the dividing unit 200B, the calculation unit 202B analyzes the halftone dots used in the image region and represents the difference in geometric properties The relationship between the quantities is determined for each region. That is, when the plane data of the binary digital image (1bit TIFF image) used to create the reference profile is input, the calculation unit 202B calculates a feature amount representing a difference in the geometric properties of halftone dots from the plane data. The relationship (first relationship) is calculated. These processes are performed in advance and the results are stored in the storage unit 201B. In the above description, although it is for each predetermined region, in this case, not a region but a relationship between feature quantities representing a difference in geometric features corresponding to the area ratio is calculated. This can be thought of as printing a test pattern having a known area ratio for each area and storing the boundary length corresponding to each area (ie, area ratio). Further, when plate data used for actual printing is input for each predetermined region, the calculation unit 202B calculates a relationship (second relationship) between the feature amounts for each region from this data. For example, the calculation unit 202B calculates, as the first relationship, a relational expression between the dot area ratio and the boundary length for the binary digital image (1bit TIFF image) used to create the reference profile, and this relational expression Is output to the storage unit 201B and stored in the storage unit 201B. When the first relationship is data, the calculation unit 202B creates a table representing these relationships and stores the table in the storage unit 201B. Further, the calculation unit 202B calculates, as a second relationship, a relational expression between the halftone dot area ratio and the boundary length for the image data (1 bit TIFF image) of the plate used for actual printing. Then, the calculation unit 202B reads the data (relational expression or table) related to the first relationship and the Neugebauer expression or the extended Neugebauer expression from the storage unit 201B, and obtains the difference between the first relationship and the second relationship. Then, as described above, the correction value is calculated using this difference and the Neugebauer equation or the extended Neugebauer equation. The calculation unit 202B outputs this correction value to the correction unit 207B.

  As described above, the storage unit 201B stores the formula or data output from the calculation unit 202B. In addition, the storage unit 201B stores a reference profile (one reference profile) for performing color space conversion by the conversion unit 206B. Further, as will be described later, the storage unit 201B may store a plurality of profiles.

  The image processing unit 205B performs the same processing as the image processing unit 205A of the image conversion apparatus 20A.

  The conversion unit 206B reads a reference profile from the storage unit 201B, performs profile conversion (color space conversion) on the CMYK multilevel image output from the image processing unit 205B using the reference profile, and outputs an RGBIr image to the correction unit 207B. To do.

  When the RGBIr image is input from the conversion unit 206B, the correction unit 207B uses the correction value (correction value calculated from the first relationship and the second relationship) input from the calculation unit 202B, and converts the image into the image. to correct. The correction unit 207B outputs the corrected image to the color tone monitoring device 21.

  FIG. 9 is a flowchart for explaining the operation of the image conversion apparatus according to the second embodiment of the present invention.

  As a pre-process, the image conversion apparatus 20B acquires the image data used to create the reference profile stored in the storage unit 201B from the image server 13 (s41). The dividing unit 200B divides the image area in the image data used for creating the reference profile into a plurality of areas (predetermined areas) for each image block, and outputs the result to the calculating unit 202B. The calculating unit 202B calculates a relational expression (first relation) between the area ratio of the halftone dots and the boundary length from the binary image data used to create the reference profile (s42). Then, the calculation unit 202B stores this relational expression in the storage unit 201B (s43).

  The image conversion apparatus 20B obtains actual print image data (1 bit TIFF) from the image server 13 before the actual printing is started (s51). The dividing unit 200B divides the image area in the image data of the reference image into a plurality of areas (predetermined areas) for each image block, and outputs the result to the calculating unit 202B. The calculation unit 202B calculates a relational expression (second relation) between the area ratio of the halftone dots and the boundary length for each predetermined area (for example, for each unit area) (s52). Further, the calculation unit 202B reads the first relational expression from the storage unit 201B, obtains a difference from the first relational expression and the second relational expression, and calculates this difference and the Neugebauer expression or the extended Neugebauer expression. The correction value (correction coefficient) is calculated by using this. Then, the calculation unit 202B outputs this correction value to the correction unit 207B (s53).

  On the other hand, the image processing unit 205B multivalues (descreens) the actual print image data (binary digital image (CMYK 1-bit image)), and outputs the CMYK multivalued image to the conversion unit 206B (s54). ).

  The conversion unit 206B reads the reference profile from the storage unit 201B, performs profile conversion (color space conversion) on the CMYK multilevel image output from the image processing unit 205B to an RGBIr reference image, and outputs the converted image to the correction unit 207B (s55). .

  The correction unit 207B corrects the image (RGBIr image) profile-converted by the conversion unit 206B with the correction value output from the calculation unit 202B, and outputs the corrected image to the color tone monitoring device 21 as a reference image (s56).

  As described with reference to FIG. 7, the color tone monitoring device 21 repeats the processes of steps s11 to s13.

  In the above description of the second embodiment, the case where only one reference profile is used has been described. However, the present invention is not limited to this. For example, a plurality of profiles may be stored in the storage unit 201B. At this time, for the image used for creating each profile by the calculation unit 202B, a relationship (first relationship) between feature amounts representing a difference in geometric properties of halftone dots is obtained and stored in the storage unit 201B. Remember. Also for the image to be actually printed, the calculation unit 202B calculates the relationship (second relationship) between the feature amounts representing the difference in the geometric properties of the halftone dots. Then, the correction unit 207B selects the profile with the smallest correction amount of the second relationship, that is, the first correction value (correction coefficient) calculated from the first relationship and the second relationship is the smallest. The relationship (relational expression) is selected. As a profile selection method, for example, the method described based on FIG. 4A in the first embodiment may be used. With this configuration, the accuracy of the reference image can be further improved.

  Note that when the correction unit 207B selects one profile from a plurality of profiles, the correction unit 207B outputs information on the profile to the conversion unit 206B. Upon obtaining this information, the conversion unit 206B reads the corresponding profile from the storage unit 201B, and performs profile conversion of the multi-valued image input from the image processing unit 205B.

  Note that a profile selected from one or more profiles may be corrected based on the relationship between the feature quantities, and an image that has undergone color space conversion may be corrected based on the corrected profile.

[Third Embodiment]
Next, a characteristic process performed by the image conversion apparatus according to the third embodiment of the present invention will be described, and then a specific configuration and process of the image conversion apparatus will be described. In the following description, the image conversion apparatus according to the third embodiment is referred to as an image conversion apparatus 20C.

  In the image conversion apparatus 20C, a profile is not prepared, and a function or LUT representing the relationship between the dot area ratio, the boundary length, the dot gain, or the effective area ratio is applied to the binary digital image, for example. The effective area ratio is taken into consideration, and a well-known Neugebauer equation is applied to this effective area ratio. It is also possible to compose the RGBIr reference image by multiplying this by the reflectance of each spectrum of the four colors of solid multiplication (solid portion and overlapping solid portion 16 colors).

  Specifically, the image conversion apparatus 20C estimates the actual area ratio (effective area ratio) at this time from a function (relational expression) representing the relationship between the area ratio of the halftone dots and the boundary length of the binary digital image, for example. To do. Then, this area ratio CMYK and the reflectivity Dk (R), Dc (R),... Dkcmy corresponding to the R, G, B, Ir light sources of each solid part and the superposed part obtained in advance by a printing test. (R), Dk (G), Dc (G), ... Dkcmy (G), Dk (B), Dc (B), ... Dkcmy (B), Dk (Ir), Dc (Ir), ... Dkcmy (Ir) is substituted into a well-known Neugebauer equation or extended Neugebauer equation and converted to a pixel value of the reference image. In the above reflectance, expressions such as k, c,... Kcmy indicate combinations of CMYK.

  From this area ratio and the boundary length, the actual area ratio (on the paper) at this time can be estimated by using a two-dimensional LUT (digital area ratio and boundary length as two variables) experimentally (print test). Is created by a method such as creating a table that outputs the estimated actual area ratio). For example, a printing plate is prepared by writing several types of CMYK test patterns in 10% increments from 0 to 100%, from fine dots to rough ones, and printing is performed using this printing plate at a standard density. At this time, the area ratio and boundary length of the digital image are input, and the actual area ratio read from the printed paper is used as an output value. A numerical value is obtained by interpolation between these grid points and registered in the LUT. The digital area ratio and boundary length are input as two variables, and the two-dimensional LUT (or function) obtained here is subtracted to output an estimated actual area ratio. For the LUT that does not have the above two-input values on the grid points, interpolation may be performed to obtain the estimated actual area ratio.

  It is also possible to create an LUT from a result of a test performed on a typical printing machine in advance without performing a print test each time, and use this.

  Further, assuming that the periphery of the boundary length is evenly expanded, an area obtained by multiplying the boundary length by an appropriate proportionality coefficient K and adding a linear function to the original area can be used as the actual area ratio. At this time, since the enlarged ones may actually overlap, a more accurate value can be obtained by approximation with a quadratic function or a cubic function.

  Next, the configuration and operation of the image conversion apparatus according to the third embodiment of the present invention will be described. FIG. 10 is a block diagram showing a schematic configuration of an image conversion apparatus according to the third embodiment of the present invention. As shown in FIG. 10, the image conversion apparatus 20C includes a dividing unit 200C, a storage unit (storage unit) 201C, a calculation unit 202C, an image processing unit 205C, and a conversion unit (image conversion unit) 206C.

  When the CMYK binary digital image data (1 bit TIFF) is input from the image server 13, the dividing unit 200 </ b> C divides the image area included in the binary digital image into a plurality of predetermined areas.

  The storage unit 201C stores an LUT or a function for converting dot ratios or effective area ratios from the reflectance of the solid part and the overlapping solid part of the printed matter and the area ratio and the boundary length obtained in advance by the print test.

  When the CMYK binary digital image data (1 bit TIFF) is input for each predetermined area from the dividing unit 200C, the calculation unit 202C analyzes a halftone dot used for the image area in the binary digital image, and A parameter representing the difference in geometric properties is calculated. As the feature amount, for example, a relational expression between the dot area ratio and the boundary length is calculated, and the dot gain or the effective area ratio of the predetermined area in the printed matter is calculated based on the LUT or function read from the storage unit 201C. When the dot gain is calculated, the effective area ratio can be obtained by applying the dot gain to the original area. If these parameters are not found in the LUT lattice points, they are obtained by interpolation. The calculation unit 202C outputs the calculation result to the conversion unit 206C.

  When the CMYK binary digital image data (1 bit TIFF) is input for each predetermined area from the dividing unit 200C, the image processing unit 205C descreens (multi-values) these binary digital images. Then, the image processing unit 205C outputs the multi-valued image subjected to the above processing to the conversion unit 206C.

  When the calculation result (dot gain) output from the calculation unit 202C is input to the conversion unit 206C, the conversion unit 206C applies dot gain to the multi-valued image input from the image processing unit 205C to obtain an effective area ratio from the storage unit 201C. Using the reflectance of the solid portion and the overlapped solid portion of the printed matter obtained in advance by the print test, the values are converted into RGBIr values using a well-known Neugebauer equation or extended Neugebauer equation, and this is output as a (reference) RGBIr image.

  When the effective area ratio is calculated by the calculation unit, the conversion unit 206C uses the reflectance of the solid portion and the overlapped solid portion of the printed matter obtained in advance by a print test for this value, and the well-known Neugebauer equation or the extended The value is converted into RGBIr values using the Neugebauer equation and output as a (reference) RGBIr image. Furthermore, by reading the density (reflectance) information of the solid portion and the overlapped solid portion of the printed matter read from the storage unit 201C, the reflectance by each spectrum of the four colors of ink (16 colors) is applied to this. The RGBIr reference image is synthesized using a predetermined model. As the predetermined model, for example, a printing machine model described in Japanese Patent No. 4176815 may be used.

  FIG. 11 is a flowchart for explaining the operation of the image conversion apparatus according to the third embodiment of the present invention. FIG. 12 is a graph showing the relationship between dot gain before and after correction at the time of multilevel (descreening).

  When the test printing is performed as a pre-process (s71), the image conversion apparatus 20C captures the continuous paper P with the optical sensors 25F and 25R, acquires the image data, and performs doting at the time of multi-value (descreening) A relationship (dot gain correction information) between the digital image and the sensor image in a single color of gain is calculated (s72). This relational expression is as shown in FIG. In the present invention, as shown in FIG. 12, a curve at the time of multi-value (descreening) can be obtained in advance by experiments or the like, and the curve can be controlled by the feature amount obtained by the calculation unit 202C. In addition, the Yule-Nielsen equation or the like can be used for these functions.

  Also, the image conversion apparatus 20C measures the RGBIr reflectance of the solid portion and the overlapped solid portion using the acquired image data (s73). Then, the calculation unit 202C stores this relational expression in the storage unit 201C (s74).

  When the actual printing is started and the image data (1bitTIFF) is received from the image server 13 (s81), the image conversion device 20C divides the image into a plurality of predetermined areas, and the calculation unit 202C performs a predetermined area for each predetermined area. Then, a relational expression between the dot area ratio and the boundary length is calculated (s82). In addition, the calculation unit 202C calculates the effective area ratio of a predetermined region in the printed material based on the LUT or function read from the storage unit 201. Then, the calculation unit 202C outputs the calculation result to the conversion unit 206C (s83).

  For each predetermined region, the conversion unit 206C converts the multivalued image obtained by the image processing unit 205C into a multivalued image based on the effective area ratio obtained in step s83 and the reflectivity of the solid part and the overlapping solid part. An RGBIr value is converted by the equation or the extended Neugebauer equation, and an estimated reference image is calculated and output using this (S84).

  As described with reference to FIG. 7, the color tone monitoring device 21 repeats the processes of steps s11 to s13.

[Others]
In the first embodiment and the second embodiment, in the image conversion apparatus, the configuration in which the profile is selected based on the result calculated by the calculation unit or the image after profile conversion is corrected has been described. It is also possible to configure so that this correction is performed before descreening.

  With this configuration, a device subsequent to the image conversion device is not a multi-value image, but a device that handles a binary image (for example, a plate making device for a rotary press having other characteristics, or a calibration device that accepts a binary image) It is effective when outputting to When outputting to a device that accepts a binary image, it is possible to prevent degradation of the image and the like due to once converting the image data converted into multiple values into a binary image again by RIP or the like.

  A configuration and processing method for outputting a processed image to a calibration device that is a device that handles a binary image by performing profile selection and image correction directly on a binary image without performing descreening will be described below. explain.

  First, the case where profile selection is performed before descreening will be described. FIG. 13 is a block diagram illustrating a configuration of the image conversion apparatus when a calibration apparatus is connected. As shown in FIG. 13A, the image conversion apparatus 20A2 includes a dividing unit (dividing unit) 200A, a storage unit (storage unit) 201A, a calculating unit (calculating unit) 202A, a determining unit 203A2, and a halftone operation unit ( Conversion means) 204A. A calibration device 20K is connected to the halftone dot operation unit 204A. Since each part of the image conversion apparatus 20A2 operates in the same manner as the image conversion apparatus 20A, only the configuration and operation different from those of the image conversion apparatus 20A will be described below.

  The determination unit 203A2 confirms the relationship between the feature amounts obtained by the calculation unit 202A, the threshold function read from the storage unit 201A, and the magnitude relationship, and determines each of the divided image regions (predetermined regions). Determine the type of screening. Then, the determination unit 203A2 aggregates the determination results and determines the most frequent screening type as the screening type for the image region. Then, the determination unit 203A2 determines whether it corresponds to the screening determined by the calibration device 20K, and outputs the information to the halftone operation unit 204A if it corresponds. If not, the determination unit 203A2 reads the screening profile and the screening profile supported by the calibration apparatus 20K from the storage unit 201A. Subsequently, an increase / decrease value of the area is obtained from the two read profiles so that the halftone dot operation unit 204A can perform halftone dot processing on the binary image, and is output to the halftone dot operation unit 204A.

  When the halftone dot operation unit 204A receives information from the determination unit 203A2 that the calibration device 20K is compatible with the screening used for the image, the halftone dot operation unit 204A does not process the binary image directly to the calibration device 20K. Output. On the other hand, when an area increase / decrease value is input from the determination unit 203A2, the halftone processing unit 204A performs a halftone processing using the area increase / decrease value, and performs the halftone processing. The image is output to the calibration device 20K. Here, the halftone dot processing is a process of expanding or contracting the periphery of each halftone dot of each CMYK binary digital image data (1 bit TIFF) input from the image server 13. As described above, the halftone processing is to expand or contract the periphery of the halftone, and can be said to be color space conversion.

  The calibration device 20K processes the binary digital image output from the halftone dot operation unit 204A, and displays or prints the image used for calibration.

  For example, screening 1 and screening 2 are used for an image input from the image server 13, and the calibration apparatus 20K supports screening 1, but does not support screening 2. The determination unit 203A2 and the halftone dot operation unit 204A perform the following processing.

  When the determined screening type is screening 1, the determination unit 203A2 outputs the fact that the screening type is screening 1 to the halftone dot operation unit 204A because the calibration device 20K supports it.

  On the other hand, the determination unit 203A2 reads the profile 1 corresponding to the screening 1 and the profile 2 corresponding to the screening 2 from the storage unit 201 because the calibration device 20K does not support when the determined screening type is the screening 2. The increase / decrease value of the area is determined for each predetermined region so that halftone dot processing of screening 2 can be performed using profile 1. That is, the area ratio 2 of the screening 2 that gives the same output as the output value of the profile 1 corresponding to the area ratio 1 calculated by the calculation unit 202A for the screening 1 is obtained, and the difference between the area ratio 1 and the area ratio 2 is predetermined. The product of the area of the region, that is, the increase / decrease value of the area is obtained. Then, the determination unit 203A2 outputs the obtained area increase / decrease value to the halftone operation unit 204A.

  When the halftone dot information of the predetermined area acquired from the determination unit 203A2 is screening 1, the halftone dot operation unit 204A sends the binary digital image input from the division unit 200A to the calibration device 20K as it is without being processed. Output.

  On the other hand, when an area increase / decrease value is input from the determination unit 203A2, the halftone dot operation unit 204A adds a pixel corresponding to the area increase / decrease value to the halftone dot for each predetermined region of the image input from the dividing unit 200A. Halftone processing is performed on the image after addition or deletion, and the processed binary digital image is output to the calibration device 20K.

  In addition, since the boundary length also changes by increasing the area, strictly speaking, it is possible to further correct this area by the increased or decreased boundary length, but in reality, when the increase or decrease of the area is not so large, Approximate increase / decrease of the boundary length can be ignored.

  Similar to the image conversion apparatus 20A illustrated in FIG. 6, the image conversion apparatus 20A2 may include an image processing unit 205A and a conversion unit 206A. In this case, the image processing unit 205A and the conversion unit 206A are connected to the subsequent stage of the halftone dot operation unit 204A, and the output of the conversion unit 206A is output to the color tone monitoring device 21. Further, in this case, the conversion unit 206A reads the reference profile from the storage unit 201A, performs profile conversion (color space conversion) on the image for each predetermined area output by the image processing unit 205A, and outputs an RGBIr image. Configure.

  Next, in the second embodiment, a case will be described in which image correction is performed directly on a binary image without performing descreening. An image conversion apparatus 20B2 illustrated in FIG. 13B includes a division unit (division unit) 200B, a storage unit (storage unit) 201B, a calculation unit (calculation unit) 202B2, and a halftone operation unit (conversion unit) 204B. Yes. Since each part of the image conversion apparatus 20B2 operates in the same manner as the image conversion apparatus 20B, only the configuration and operation different from those of the image conversion apparatus 20B will be described below.

  The calculation unit 202B2 includes a relationship (first relationship) between feature amounts representing a difference in geometric properties of the binary digital image used for creating the reference profile, and image data of a plate used for actual printing. And calculating a relationship (second relationship) between the feature quantities representing a difference in geometric property about the difference between the first relationship and the second relationship, and calculating the difference between the difference and the predetermined area Is output to the halftone dot operation unit 204B as an increase / decrease value.

  When the increase / decrease value is input from the calculation unit 202B2, the halftone dot operation unit 204B adds or deletes pixels corresponding to the increase / decrease value to the halftone dot for each predetermined region of the image input from the division unit 200B. The binary digital image subjected to the point processing is output to the calibration device 20K.

  Similar to the image processing unit 20B illustrated in FIG. 8, the image conversion device 20B2 may include an image processing unit 205B and a conversion unit 206B. In this case, the image processing unit 205B and the conversion unit 206B are connected downstream of the halftone dot operation unit 204A, the correction unit 207B is connected downstream of the calculation unit 202B, and the output of the conversion unit 206B is output to the color tone monitoring device 21. Configure as follows. In this case, the correction unit 207B is not necessary.

  In the image conversion apparatus 20A and the image conversion apparatus 20B shown in FIG. 13, the increase / decrease of the halftone dot obtains the boundary length (number of pixels) of a predetermined area, and obtains the ratio between the number of pixels to be increased / decreased and the number of the boundary length pixels. . Furthermore, when a predetermined area is raster-scanned and a boundary point is found, the value of the pixel at the boundary point is changed from white to black in the case of increase and from black to white in the case of decrease according to the above ratio. Realize.

  For example, when increasing the number of pixels by 100 in a predetermined area, if the boundary length is 400 pixels, raster scanning is performed, and when a boundary point is found, white may be changed to black once every four times.

  In this way, the binary image is increased / decreased (expanded / contracted) at a rate corresponding to the calculated result and output. Then, a printing plate is created from the binary image by CTP, for example, and printed. As a result, almost the same print result can be obtained for different screening.

  With this configuration, the calibration device 20K is not normally compatible with various types of screening, but the image conversion device 20D outputs an image that has been subjected to halftone operation according to various types of screening to the calibration device 20K. Incorrect display and printing can be prevented by the calibration device 20K.

  In the above description, as a method for color space conversion of a binary digital image, a known method may be used in which a binary digital image (CMYK) is multi-valued and further converted into an RGBIr image by color space conversion. This method may be used. For example, RGBIr values corresponding to 16 patterns of this combination are obtained from a binary table (CMYK) in the upstream process from a pre-registered table. Then, by integrating this within the range of the effective visual field of one pixel of the optical sensor, a provisional RGBIr value for each pixel of the optical sensor is obtained. Also, an appropriate test pattern is printed, and the relationship between the provisional RGBIr value at this time and the RGBIr value read by the optical sensor when actually being properly printed is obtained as a function LUT having four variables and four outputs. Keep it. In actual use, a provisional RGBIr value is obtained from a CMYK binary digital image, and the above function or LUT is applied to the provisional RGBIr value to calculate a reference RGBIr value. Subsequently, the reference RGBIr value is compared with the RGBIr value read by the sensor, and the opening of the ink key of the printing press is controlled so that the two match, thereby adjusting the color tone of the printed matter to a desired state. be able to.

  Each of the image conversion apparatuses 20 shown in FIGS. 6, 8, 10, and 13 may be configured by hardware, or may be configured by software (program) using a computer. . For example, when the image conversion apparatus 20A shown in FIG. 6 is configured by a computer, the control unit reads the image conversion program from the storage unit 201A and executes the program, so that the computer executes the storage unit 201A and the calculation unit 202A. -It is good to comprise so that it may function as the determination part 203A, the image process part 205A, and the conversion part 206A. The image conversion apparatus 20 shown in other figures can be replaced in the same manner.

    DESCRIPTION OF SYMBOLS 1 ... Printing system 13 ... Image server 17 ... CTP 19 ... Rotary press 20 ... Image converter 20K ... Calibration apparatus 21 ... Color tone monitoring apparatus 22 ... Monitor 23 ... Ink key control apparatus 25F, 25R ... Optical sensor 72A-72H ... Ink key 73 ... Printing unit

Claims (8)

A calculation means for obtaining a relationship between two or more different feature amounts among a plurality of feature amounts representing a difference in geometric properties of the halftone dots for screening of a binary digital image;
Conversion means for performing color space conversion of the binary digital image based on the relationship between the feature amounts calculated by the calculation means;
An image conversion apparatus comprising :
The calculation means obtains a relationship between at least two feature amounts among a halftone dot area ratio, a boundary length of a binary image forming a halftone dot, a halftone dot area, the number of halftone dots, and a spatial frequency characteristic. Means,
The converting means includes
A determination function for determining the type of screening, and storage means for storing a profile;
A determination unit that determines a type of screening based on the determination function and a relationship between the feature amounts calculated by the calculation unit;
Image conversion means for reading a profile corresponding to the determination result of the determination means from the storage means, and performing color space conversion of the binary digital image based on the profile;
An image conversion apparatus. The image conversion apparatus according to claim 1 , further comprising a dividing unit that divides the binary digital image into a plurality of predetermined regions and outputs the divided digital image to the calculation unit and the conversion unit. The dividing means separates the binary digital image into a plurality of image blocks, further divides each image block into a plurality of regions,
The calculation means obtains a relationship between feature amounts for each area divided by the division means,
The determination means determines a screening type for each region, totals the determination results for each image block, and determines a screening type for the image block based on an evaluation function for evaluating the total results. The image conversion apparatus according to claim 1 or 2 . In the case where the binary digital image is composed of a plurality of plates, the determination unit determines the type of screening for each of the plurality of predetermined areas for each plate, and further aggregates the determination results for each plate. an image conversion apparatus according to any one of claims 1 to 3 to determine the type of screening of the binary digital images. A storage means for storing the profile;
The calculation means obtains a first relationship between feature quantities representing a difference in the geometrical properties of halftone dots for the binary digital image used to create the profile, and uses the binary digital image used in actual printing. For a second relationship between features representing the difference in geometric properties of the halftone dots,
The converting means includes
Image conversion means for reading out the profile from the storage means and performing color space conversion of the binary digital image based on the profile;
A correction unit that calculates a correction value from the first relationship and the second relationship and corrects an image that has been subjected to color space conversion by the image conversion unit based on the correction value;
An image conversion apparatus according to claim 1, comprising: The converting means includes
Storage means for storing the reflectance information of each color of the solid portion and the overlapped solid portion of the printed material, and dot gain correction information;
The relationship between the feature values obtained by the calculation means, the density information of the solid portion and the overlapped solid portion of the printed matter read from the storage means, the correction information of the dot gain, the well-known Neugebauer formula or the extended Neugebauer formula , Based on the image conversion means for performing color space conversion,
An image conversion apparatus according to claim 1, comprising: The conversion means performs a color space conversion for expanding or contracting halftone dots of binary digital image data based on the relationship between the feature amounts calculated by the calculation means, and a halftone operation for outputting the processing result to the outside image converter according to any one of claims 1 to 6 including means. Computer
A calculation means for obtaining a relationship between two different feature quantities among the feature quantities representing a difference in geometric properties of halftone dots for screening of a binary digital image input from the outside;
Conversion means for performing color space conversion of the binary digital image based on the relationship between the feature values obtained by the calculation means;
An image conversion program to function as,
The calculation means obtains a relationship between at least two feature amounts among a halftone dot area ratio, a boundary length of a binary image forming a halftone dot, a halftone dot area, the number of halftone dots, and a spatial frequency characteristic. Means,
The converting means includes
A determination function for determining the type of screening, and storage means for storing a profile;
A determination unit that determines a type of screening based on the determination function and a relationship between the feature amounts calculated by the calculation unit;
Image conversion means for reading a profile corresponding to the determination result of the determination means from the storage means, and performing color space conversion of the binary digital image based on the profile;
An image conversion program which is means comprising:

Images