JP4212822B2 - Image recording apparatus and gradation correction data creation method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image recording apparatus such as a copying machine that reads a document image and records an image corresponding to the document image on a sheet, and more particularly to signal processing for faithfully recording a halftone on a document.
[0002]
[Prior art]
When recording a halftone with an image recording apparatus such as a copying machine or a printer, there is a problem that the density of the halftone differs from machine to machine or changes over time. In order to solve this problem, a plurality of gradation patterns are recorded in advance, optically read by a scanner or the like, the gradation characteristics of the recording device are calculated, and correction data is created based on this. In other words, means for performing gradation correction using this correction data during image recording has been used. Further, the interpolation data is obtained by approximating with one high-order function based on the reading values of the gradation patterns in a plurality of stages, and the interpolation data is created based on this approximation function.
[0003]
[Problems to be solved by the invention]
The types of screen patterns are increasing due to the addition of a printer function to a plurality of document modes and PPC. Although it is necessary to perform gradation correction for each screen pattern, if gradation correction is performed for various screen patterns recorded on a plurality of sheets, the operation takes time and becomes complicated.
[0004]
By outputting a plurality of screen patterns on a single sheet, the above-described complexity is solved, but the number of patches per screen pattern is reduced due to sheet size restrictions, which may adversely affect the correction accuracy.
[0005]
Approximating the gradation characteristics for all gradation areas with a high-order polynomial requires, for example, a higher order of 8th order or more, resulting in an enormous amount of calculation, a large error due to digit loss, and a decrease in correction accuracy. There was a problem.
[0006]
Accordingly, an object of the present invention is to eliminate the complexity of gradation correction processing when using a plurality of screen patterns, and to create highly accurate gradation correction data using a small number of gradation patterns.
[0007]
[Means for Solving the Problems]
In order to achieve the above object, a gradation correction data creation method according to an embodiment of the present invention uses an image reading unit that optically reads an image and outputs an image signal, and uses the gradation correction data as the image signal. A gradation correction processing unit for correcting the image data and an image recording apparatus including an image recording unit for recording an image according to the corrected image signal. A step of recording a predetermined screen pattern including a plurality of types of screen patterns on one sheet using the image recording unit, and reading the recorded predetermined screen pattern by the image reading unit, Providing a screen pattern reading signal; reading data values of the patch corresponding to the plurality of types of screen patterns; and each screen pattern. By adding the standard gradation characteristic of each screen pattern to one approximate function for the whole of a plurality of types of screen patterns obtained from the difference from the standard gradation characteristic value of each screen pattern, each screen pattern Interpolating data corresponding to, and creating the gradation correction data based on the interpolating data.
[0009]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described in detail with reference to the drawings. The following description is an embodiment of the present invention and does not limit the apparatus and method of the present invention.
[0010]
<First Embodiment>
FIG. 1 is a block diagram showing a configuration of an image forming apparatus according to an embodiment of the present invention.
[0011]
The image input unit 101 optically reads a document image and outputs it as an RGB signal 151. The signal integration unit 102 integrates the read RGB signal 151 in a designated area. The color conversion unit 103 converts the RGB signal 151 into a CMY signal 152 representing the density of each color of the printer. The image area identification unit 104 receives the RGB signal 151 and the CMY signal 152, identifies the character area, the gradation area, and the like using information such as the edge structure, saturation, and density, and outputs an identification signal 153. The filter unit 105 filters the CMY signal 152. The filter unit 105 switches the filter characteristics in accordance with the image area signal 153 output from the image area identification unit 104.
[0012]
The inking unit 106 adds a black plate (K plate) signal to the CMY signal 154 output from the filter unit 105 and outputs a CMYK signal 155. As a black plate calculation method, replacement with one black color, a GCR (gray component removal) method, or the like is selected by an identification signal 153. If the identification result is a black character area, black color processing is selected to express black characters clearly, and GCR is selected in other areas to improve gradation reproducibility and color reproducibility.
[0013]
The tone correction unit 107 outputs a CMYK signal 156 by performing gamma correction on the signal value using the LUT so that the tone characteristics of the tone processing unit 108 and the image recording unit 109 are linear with respect to the value of the signal 155. . The gradation processing unit 108 performs screen processing on the CMYK signal 156 to perform area modulation, and outputs a recording signal 157. In the test mode, the pattern generator 112 outputs later-described gradation patterns corresponding to various image areas and screen patterns.
[0014]
The image recording unit 109 records an image on the sheet according to the image signal 157. The interface unit 110 is connected to an external general-purpose computer or a dedicated RIP machine, inputs an image signal 158 from the outside, and supplies it to the gradation correction unit.
[0015]
Next, gradation correction and gradation processing according to the present invention will be described.
[0016]
In electrophotographic recording, if density is expressed for each pixel, it tends to be unstable. In many cases, gradation is expressed by an area modulation method. The area modulation method includes a method using a regular pattern repetition and a method using an irregular pattern, and the regular pattern is called a screen. Here, the screen generally indicates a set of regular patterns.
[0017]
As shown in FIG. 2 (a), the screen pattern is composed of parallel lines at equal intervals and expresses shades by the thickness of the parallel lines, or two-dimensional regular points as shown in FIG. 2 (b). There are a halftone dot pattern that is composed of a periodic structure and expresses shading according to the size of a point, or a method of switching between a line and a halftone dot according to the density. As a method using an irregular pattern, error diffusion, blue noise mask, and the like are known.
[0018]
In an example of the line pattern shown in FIG. 2A, signal values of three horizontal pixels are added, and a line image having a width corresponding to the added value is formed at the center of the three pixels. Further, in the case of FIG. 2A, the pattern is shifted by one pixel every two scanning lines. The resulting gradation image is an image composed of a plurality of diagonal lines.
[0019]
In an example of the halftone dot pattern shown in FIG. 2B, the signal values of 4 horizontal pixels are added, and an image having a width corresponding to the added value is formed at the center of the 4 pixels. Further, in the case of FIG. 2B, the pattern is shifted by two pixels for every two scanning lines. The gradation image obtained as a result is an image composed of independent points.
[0020]
Hereinafter, the present embodiment will be described by taking as an example the case where the line method is used as the screen processing.
[0021]
Image quality characteristics such as gradation reproducibility and resolution change depending on the screen pattern density. In the case of the multi-line method, if the density is increased (the interval between adjacent parallel lines is reduced), the resolution is increased and fine portions of the image can be expressed, but the gradation stability, that is, the gradation reproducibility is lowered. On the other hand, when the density is decreased (the interval is increased), the resolution is lowered and a line of texture can be seen, but the gradation stability is improved.
[0022]
A screen with a high density is suitable for recording images such as characters and line drawings, and a screen with a low density is suitable for recording gradation images such as photographs. For this reason, when the screen patterns having different densities are switched between the character area and the photograph area, both areas are well reproduced.
[0023]
In addition, the user can specify the document mode according to the type of document, and use a screen pattern suitable for the document image by changing the density when recording a photographic image and recording a character image. Can do.
[0024]
As a signal processing method for generating such a screen, there are methods such as a multi-value dither method and a direct pattern selection method. Screen processing can be performed by referring to tables and performing calculations according to these methods.
[0025]
In electrophotographic recording, the recording unit generally has a nonlinear characteristic between the input image signal value and the output image density. Furthermore, the characteristics fluctuate due to environmental conditions such as temperature and humidity and variations from machine to machine. The relationship between the input image signal value and the output image (printed image) density in gradation processing is called gradation characteristics. The tone correction unit 107 corrects (non-gamma correction) this non-linearity, temporal variability, and machine-to-machine variation. The gradation correction unit 107 is configured by an LUT, and converts a signal so as to reproduce a linear density according to an input image signal.
[0026]
Since the gradation characteristics differ depending on the screen pattern to be used, it is necessary to have a correction table corresponding to each screen pattern. In the case of color, since the gradation characteristics differ for each recording color, it is necessary to have a correction table for each recording color.
[0027]
FIG. 3 shows detailed configurations of the gradation correction unit 107 and the gradation processing unit 108 according to the present embodiment.
[0028]
In the normal document mode, screen patterns having different densities are used for the photo area and the character area. Therefore, the gradation correction unit 107 includes a correction table 121a and a first gradation correction unit 122a for a photographic area, and includes a correction table 121b and a second gradation correction unit 122b for a character area. A screen pattern independent of the CPU can be set in these tables.
[0029]
The gradation processing unit 108 includes a screen pattern 123a and a first gradation processing unit 124a for a photographic area, and a screen pattern 123b and a first gradation processing unit 124b for a character area. The outputs of the two gradation processing units 124a and 124b are selected by the signal selection unit 125 according to the identification signal.
[0030]
Further, since the screen pattern is changed depending on the recording color, gradation correction and gradation processing are performed on the photographic area and the character area for each of the four recording colors of CMYK.
[0031]
The gradation correction unit 107 performs density characteristic correction (gamma correction). The correction is performed using an LUT (Look Up Table). Since the gradation characteristics are different for each screen pattern, this embodiment has a total of eight correction tables of 4 colors x 2 of CMYK.
[0032]
FIG. 4 is a flowchart showing the operation in the test mode according to the present embodiment.
[0033]
The pattern generator 112 generates a predetermined screen pattern composed of a plurality of gradation patches as gradation test pattern data (ST101). Each patch is, for example, a 10 mm square uniform gradation pattern. The gradation test pattern data generated from the pattern generation unit 112 is screen-processed by the gradation processing unit 108 and recorded on one sheet by the image recording unit 109 (ST102).
[0034]
Since it is difficult to record all the gradation patches on one sheet by the image recording unit 109, a patch having a predetermined number of gradation values obtained by thinning out all gradation values is recorded as a gradation test pattern. . The characteristics of all gradation values are calculated by an interpolation process described later.
[0035]
An example of the gradation test pattern of this embodiment is shown in FIG.
[0036]
The black bar 401 is for position detection and is used to specify the position of the patch. Below the black bar 401, there are 8 rows and 10 rows of photographic area screen patch areas 402, and there are 8 rows and 10 rows of character area screen areas 403 below.
[0037]
C1, C2,... C20 are patches for a C (cyan) color photographic region, C1 is the patch with the lowest density (white), and C20 is the patch with the highest density (solid). Similarly, M1 to M20, Y1 to Y20, and K1 to K20 are M (magenta), Y (yellow), and K (black) patches for photography, respectively. Similarly, C21 to C40, M21 to M40, Y21 to Y40, and K21 to K40 are C, M, Y, and K character area patches, and C21, M21, Y21, and K21 are the lowest densities (white), and C40, M40. , Y40, K40 is the highest density (solid).
[0038]
The image input unit 101 reads the gradation test pattern recorded on the paper (ST103). The signal integration unit 102 integrates a plurality of pixel signals. That is, the signal integration unit 102 integrates pixel signal values in the area corresponding to each patch in the input image area. The CPU 111 reads the integrated value.
[0039]
The CPU 111 performs the following calculation and creates a gradation correction table. Here, a cyan photographic region pattern will be described. The reading values of the 20 patches C1 to C20 for the cyan photographic region are YC1 to YC20. Among the RGB signals, an R signal that is a complementary color component of cyan is used. The R signal is used here because the R component light receiving element of the image input unit 101 has the highest sensitivity to cyan among the R, G, and B component light receiving elements, and a high signal S / N is obtained.
[0040]
The CPU 111 performs standardization using the following formula (ST104).
[0041]
ZCi = (YCi-YC1) / (YC20-YC1)
ZCi is 1 when YCi is the maximum density (YC20), and 0 when the minimum density (YC1). Thus, each YCi is normalized to ZCi. This ZCi is called area ratio.
[0042]
FIG. 6 is an explanatory diagram for creating a gamma correction table.
[0043]
The values of the signal 156 when the patches C1 to C20 are recorded are XC1 to XC20. FIG. 6A shows the area ratio ZCi with respect to the signal values XC1 to XC20. That is, FIG. 6A is a diagram in which the read values YCi (YC1 to YC20) obtained by reading the patches recorded with the signal values XC1 to XC20 are normalized and represented by the area ratio ZCi. As will be described later, the CPU 111 obtains a continuous approximate function such as a curve indicated by 502 in FIG. 6B from a point sequence 501 shown in FIG. 6A (ST105). The CPU 111 further calculates an inverse function 503 of the approximate function 502 as shown in FIG. 6C, and this becomes a gradation correction table (ST106).
[0044]
Similarly to the above, the CPU 111 sequentially creates a MYK gradation correction table for the photographic screen and a CMYK gradation correction table for the character area screen (ST107). These gradation correction tables need to be calculated a plurality of times (eight times in the case of this embodiment), but the gradation test pattern can be recorded and read only once as in steps ST102 and ST103.
[0045]
Next, details of an approximate function calculation method as in step ST105 will be described.
[0046]
First, as shown in FIG. 7A, C1 to C20 are divided into the following 17 groups A1 to A17 while allowing duplication.
[0047]
A1: C1-C4
A2: C2 to C5
A3: C3-C6
...
...
A17: C17 to C20
An approximate function Fi is calculated for each group Ai. For example, an Nth order polynomial is used for calculating the approximate function. In this embodiment, a least square method is used, and a quadratic function that minimizes the sum of squares of approximation errors is used.
[0048]
For example, in group A1, E = Σ (F1 (XCj) −ZCj) ^ 2 from four points (XC1, ZC1) to (XC4, ZC4).
1 ≦ j ≦ 4
* X ^ 2 represents the square of X.
[0049]
Calculate a quadratic function F1 that minimizes. The curve in FIG. 7B shows a quadratic function F1. Similarly, approximate functions (F2 to F17) after group A2 are also obtained.
[0050]
Since the approximate functions calculated in this way are discontinuous, weighted synthesis is performed in an overlapping region between groups, and this result is defined as a final approximate function F (x). For example,
When XC1 ≦ x ≦ XC2 F (x) = F1 (x)
When XC2 ≦ x ≦ XC3 F (x) = ((XC3-x) · F1 (x) + (x−XC2) · F2 (x)) / (XC3−XC2)
When XC3 ≦ x ≦ XC4 F (x) = ((XC4-x) · F1 (x) + (x−XC3) · F2 (x)) / (XC4-XC3)
(In this case, the F3 component is not included to simplify the formula, but of course F3 may be included)
Same as below.
In the past, for example, interpolation was performed using high-order polynomials for all the intervals to perform approximate calculations. Therefore, a higher order of 8th order or higher is required, and the amount of calculation is enormous and errors due to digits are large. There were problems such as reduced accuracy. However, by using section division as described above, it is possible to perform highly accurate interpolation with a simple calculation.
[0051]
As described above, according to the first embodiment of the present invention, even when screen patterns having different gradation characteristics are used for photographs and characters, gradation is produced by outputting (printing) one gradation test pattern. A correction table can be created. Further, the gradation correction table can be created by reading the gradation test pattern once. This reduces the complexity of the gradation correction operation. Further, by obtaining the interpolation function by dividing the patch data into groups, high correction accuracy can be obtained with a small amount of calculation.
[0052]
<Second Embodiment>
Next, a second embodiment of the present invention will be described. The block configuration is the same as in the first embodiment. In this embodiment, the screen pattern used in the printer and the screen pattern used in the copying machine are different from each other. Accordingly, the number of screen patterns and correction tables is doubled compared to the first embodiment.
[0053]
The gradation test pattern of this embodiment is shown in FIG. This gradation test pattern is roughly divided into left and right, and a screen pattern patch 702 for a printer is arranged on the left side and a screen pattern patch 703 for a copying machine is arranged on the right side.
[0054]
The calculation of the interpolation function of the present embodiment is the same as that of the first embodiment, but the section division method is different. Depending on the screen pattern, a gradation characteristic with a long flat portion region as shown in FIG. 9 may be obtained. In such a case, by suddenly taking one section in the flat portion (group A1 in FIG. 9), sudden data fluctuation is suppressed.
[0055]
According to this embodiment, even when screen patterns having different gradation characteristics are used in the printer mode and the copier mode, a gradation correction table can be created by outputting one gradation pattern. This reduces the complexity of the gradation correction operation. Further, by dividing the patch data into groups to obtain an interpolation function, and further increasing the group size in a flat gradation portion, high correction accuracy can be obtained with a small amount of calculation.
[0056]
<Third Embodiment>
Next, a second embodiment of the present invention will be described. The block configuration is the same as in the first embodiment. This embodiment has a plurality of document modes (character photo, text, photographic paper photo, and print photo), and uses different screen patterns depending on the document mode.
[0057]
The structure of the gradation pattern is shown in FIG. Screen patterns corresponding to the four document modes are arranged in the upper left (902), upper right (903), lower left (904), and lower right (905). Since many types of patterns are arranged, the number of patches in one pattern is as small as ten. Accordingly, when there are four types of document modes as in the present embodiment, 40 gradation patches are set for each color. That is, the patch shown in FIG. 10 has 160 gradation patches in total. For simplicity, only cyan is described below.
[0058]
In this embodiment, since the number of patches (the number of gradations) is small, the approximation accuracy may be lowered. In order to prevent this, gradation correction data is created using measured values of different screen patterns.
[0059]
The calculation method of the approximate function of this embodiment is demonstrated with reference to FIG.
[0060]
In this embodiment, 10 gradation patches are recorded for each document mode. Screen pattern patches for the first document mode are C1 to C10, screen pattern patches for the second document mode are C11 to C20, and patches corresponding to the third and fourth document modes are similarly C21 to C30 and C31. ˜C40. Assuming that the signal value output from the patch Ci is XCi, in this embodiment, as shown in FIGS. 11A to 11D and the following formula, the density of the patches having the corresponding density order in each mode is sequentially shifted and set. To do.
[0061]
XC1 <XC11 <XC21 <XC31 <XC2 <XC12 <...
In addition, standard gradation characteristics G1 (x), G2 (x), G3 (x), and G4 (x) of each screen pattern are measured and stored in advance. FIG. 11A shows the standard gradation characteristics G1 (x) of the first screen pattern, patch signal values XC1, XC2,..., And patch area ratios ZC1, ZC2,.
[0062]
Next, a value obtained by subtracting the standard gradation characteristic G1 (x) from the area ratios ZC1, ZC2,... Is set as QC1, QC2,. These represent variations from the standard gradation characteristics. FIG. 11E corresponds to XC1 to XC3. Variations QC1 to QC3 are shown.
[0063]
Similarly, fluctuations in the second, third, and fourth screen patterns are calculated as shown in FIGS. 11 (f) to 11 (h). When these fluctuations are superimposed, the result is as shown in FIG. By calculating an approximate function G5 (x) from these point sequences and adding the standard gradation characteristics of each screen pattern to the obtained approximate function G5 (x), an approximate function of the screen pattern can be obtained.
[0064]
The variation data as shown in FIGS. 11 (e) to 11 (h) is data relating to different screen patterns, but is highly correlated since it is data of the same color component. Therefore, if the approximate function G5 (x) is obtained from these data, the approximate calculation is performed using substantially four times the data, and the correction accuracy is improved. Also, by increasing the weight of data from the screen pattern calculated when obtaining the approximate function, the correction accuracy can be improved even when the correlation between the screen patterns is small.
[0065]
As described above, according to the third embodiment, even when screen patterns having different gradation characteristics are used in a plurality of document modes, a gradation correction table can be created by outputting one gradation pattern. This reduces the complexity of the gradation correction operation.
[0066]
Further, by creating gradation correction data using gradation data of a pattern other than the corresponding screen pattern, high correction accuracy can be obtained with a small number of patches.
[0067]
【The invention's effect】
According to the present invention, even when screen patterns having different gradation characteristics are used, a gradation correction table for each screen pattern can be created by outputting a single gradation pattern, and the operation of gradation correction is complicated. Is reduced.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a configuration of an image forming apparatus according to an embodiment of the present invention.
FIG. 2 is a diagram showing an example of a screen pattern.
FIG. 3 is a block diagram illustrating a detailed configuration of a gradation correction unit and a gradation processing unit.
FIG. 4 is a flowchart showing an operation in a test mode according to the present embodiment.
FIG. 5 is a diagram illustrating a configuration example of a gradation pattern.
FIG. 6 is a view for explaining creation of a gamma correction table.
FIG. 7 is a diagram for explaining a method of calculating an interpolation function.
FIG. 8 is a diagram illustrating a second configuration example of a gradation pattern.
FIG. 9 is a diagram for explaining long gradation characteristics of a flat portion.
FIG. 10 is a diagram showing a third configuration example of a gradation pattern.
FIG. 11 is a diagram for explaining another calculation method of an interpolation function.

Claims (3)

An image reading unit that optically reads an image and outputs an image signal, a tone correction processing unit that corrects the image signal using tone correction data, and an image that records an image according to the corrected image signal A method for creating the gradation correction data in an image recording apparatus including a recording unit,
Recording a predetermined screen pattern including a plurality of types of screen patterns each having a plurality of gradation patches on a sheet of paper using the image recording unit;
Reading the recorded predetermined screen pattern with the image reading unit and providing a read signal of the predetermined screen pattern;
One approximate function for the whole of the plurality of types of screen patterns obtained from the difference between the read data value of the patch corresponding to the plurality of types of screen patterns and the standard gradation characteristic value of each screen pattern, Adding the standard gradation characteristics of the screen pattern to create interpolation data corresponding to each screen pattern;
Creating the gradation correction data based on the interpolation data;
A method for creating gradation correction data, comprising:   6. The gradation correction data creation method according to claim 5, wherein each of the predetermined screen patterns has a plurality of gradations and a plurality of screen patterns corresponding to a plurality of document modes. An image reading unit for optically reading an image and outputting an image signal;
A gradation correction processing unit that corrects the image signal output from the reading unit using gradation correction data;
An image recording apparatus comprising an image recording unit that records an image according to a corrected image signal,
The gradation correction processing unit
Pattern generating means for generating a predetermined screen pattern including a plurality of types of screen patterns each having a plurality of gradation patches;
Recording means for recording the predetermined screen pattern generated by the pattern generating means on a sheet of paper using the image recording unit;
Pattern reading means for reading the recorded predetermined screen pattern by the image reading unit and providing a read signal of the predetermined screen pattern;
One approximate function for the whole of the plurality of types of screen patterns obtained from the difference between the read data value of the patch corresponding to the plurality of types of screen patterns and the standard gradation characteristic value of each screen pattern, Interpolation data creation means for creating interpolation data corresponding to each screen pattern by adding the standard gradation characteristics of the screen pattern of
Correction data creating means for creating the gradation correction data based on the interpolation data;
An image recording apparatus comprising:

Images