JPH04266265A - Picture correction method and picture copying machine with the method applied thereto - Google Patents

Abstract

PURPOSE:To realize always excellent copying of a picture even when the density characteristic is subject to change due to a difference of a structure between machines or an external environmental change by correcting properly the complicated reproduction density characteristic of the picture copying machine. CONSTITUTION:A CPU 112 allows a pattern generator 113 to generate a color patch based on pattern signals Yo, Mo, Co, BKo decided in advance for each of Y, M, C, Bk and a print processing section 111 prints out the patch. Feedback signals Y, M, C, Bk are obtained by applying color separation, shading correction, logarithmic transformation and masking processing similar to the picture output processing to the gradation pattern patch. The feedback signal is stored in a memory 114. The color correction is implemented so that a pattern signal and the feedback signal of a color original are identical to each other.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image correction method and an image copying apparatus, and more particularly to an image correction method for integrally correcting the image reproduction characteristics of the input and reproduction output of an original, and an image copying apparatus to which the method is applied. be.

0002

2. Description of the Related Art FIG. 6 is a block diagram showing the schematic structure of a conventional image copying apparatus. The operation of the conventional color image copying apparatus will be described below.

A color original 201 placed on a glass original table is imaged onto a line sensor 203 by an imaging lens 202 . The line sensor 203 is coated with three color filters of R, G, and B, and color separation signals of the original are output for each pixel. The output of the line sensor 203 is converted into a digital signal via a sample hold circuit 204 and an A/D conversion circuit 205, and a shading correction circuit 206 removes sensitivity unevenness, illumination unevenness, etc. of the line sensor.

[0004] In the logarithmic conversion circuit 207, the RGB signals are converted into complementary colors C (cyan), M (magenta), and Y (yellow). The output Y, M, and C signals of the logarithmic conversion circuit 207 are subjected to matrix calculation processing in a masking circuit 208 to correct unnecessary absorption of dye during print output, and then processed by a blacking circuit and a UCR (undercolor removal) circuit 209. In this step, processing is performed to replace the minimum values of Y, M, and C with Bk (black) signals. Y, M, obtained in this way,
The C and Bk signals are sent to the print processing section 211. In the print processing unit 211, the Y, M, C, and Bk signals are first binarized in a binarization circuit 601, and the binarized data is output to a driver circuit for driving a print head 602 for each color. Ru. The print head 602 scans the transfer paper 212, and the output image is reproduced and recorded. Here, as the binarization circuit 601, a dither method, an error diffusion method, etc. are considered, and as the print head 602, an inkjet method, an electrophotographic method, a thermal transfer method, a silver salt photographic method, etc. are known.

[0005]

However, in the conventional example described above, for example, in a color image copying apparatus having an inkjet type image output means, the reproduced density largely depends on the amount of ink ejected, and the amount of ink ejected depends on the ink head. This is largely due to the nozzle system. Since there are slight differences in the nozzle system of the ink head from lot to lot, the printing density also varies from lot to lot of ink heads. Further, even in a color image copying apparatus having an electrophotographic image output means, the reproduction density characteristics change in a complicated manner depending on the external environment such as temperature and humid noise. Therefore, the conventional example has the disadvantage that good color reproduction of color images cannot always be achieved due to the instability of density characteristics.

FIG. 7 illustrates an example of the variation in the density characteristics. Normally, the reproduction density characteristics of an image output device (hereinafter referred to as a printer) are set so that the printer input signal and reproduction density have a linear relationship based on the reproduction density (solid density) at the maximum printing ratio. (Figure 7
straight line ■). However, due to the above-mentioned reasons, the density characteristics of the printer fluctuate, for example, as shown by curve (2). Curve■
In comparison with the straight line (■) indicating the standard state, the reproduction characteristic is that light areas are reproduced more faintly and high-density areas are reproduced more intensely.

In order to cope with such variations in density characteristics, when the input image signal is "X", a linear conversion process (X'=kX) in which the image signal X is multiplied by a constant (k times) is performed using a color balance function. However, such a simple correction cannot cope with the above-mentioned complex variations in density characteristics. That is, for the reproduced density characteristic shown in curve
．． When the value is set to a value of 0 or more, it is impossible to correct the entire range of color reproduction density, as shown in curve (3), where the reproduced density becomes even darker in a high density area.

The present invention has been made in view of the above-mentioned problems, and provides an image correction method and an image correction method that can always realize good image copying by appropriately correcting the complicated reproduction density characteristics of an image copying device. An image copying apparatus to which the method is applied is provided.

[0009]

[Means for Solving the Problems] In order to achieve the above object, the image correction method of the present invention is an image correction method for correcting an image in an image copying apparatus that inputs an image original and obtains a reproduced image, the method comprising: A process of outputting an image in response to a predetermined gradation signal, and a gradation signal created when the outputted image is input as an image original match the predetermined gradation signal. and a step of setting a correction value based on a multidimensional functional formula.

The image copying apparatus of the present invention is an image copying apparatus that inputs an image original and obtains a reproduced image, and further includes image output means for outputting an image in accordance with a predetermined gradation signal; correction value setting means for setting a correction value based on a multidimensional function formula so that a tone signal created when the image is inputted as an image original matches the predetermined tone signal; and image correction means for correcting a newly input document image using the correction value set in the correction value setting means.

[0011] Here, the predetermined gradation signal is a discrete signal, and the correction value setting means includes interpolation means for interpolating between the correction values obtained from the discrete gradation signal. Further, the correction value setting means includes coefficient setting means for setting each dimensional coefficient of a multidimensional function equation that approximates each correction value,
The image correction means corrects the image based on the multidimensional functional expression. Further, the image correction means includes a storage means for storing correction values using the input gradation signal as an address, and the correction value setting means stores the correction values in the storage means.

In the image copying apparatus configured as described above, a reproduced image is created in accordance with a predetermined gradation signal, and this reproduced image is read as a document and is matched with the predetermined gradation signal. Since the correction value is set based on a multidimensional function formula so as to ensure that the image is copied in good quality at all times.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. Although this embodiment will be described using color image information as the image information, the image copying apparatus of this embodiment is not limited to this.

<Configuration of this Embodiment> FIG. 1 is a block diagram showing a schematic configuration of an image copying apparatus of this embodiment. Referring to FIG. 1, first, the operation until a color document placed on the document table glass is output will be described.

A color original 101 placed on a glass original table is imaged onto a line sensor 103 by an imaging lens 102 . The line sensor 103 is coated with three color filters of R, G, and B, and color separation signals of the original are output for each pixel. The output of the line sensor 103 is converted into a digital signal via a sample hold circuit 104 and an A/D conversion circuit 105, and a shading correction circuit 106 removes sensitivity unevenness, illumination unevenness, etc. of the line sensor 103.

In the logarithmic conversion circuit 107, the RGB signals are converted into complementary colors C (cyan), M (magenta), and Y (yellow). The output Y, M, and C signals of the logarithmic conversion circuit 107 are subjected to matrix calculation processing in a masking circuit 108 to correct unnecessary absorption of dye during print output, and then processed by a blacking circuit and a UCR (undercolor removal) circuit 109. In this step, processing is performed to replace the minimum values of Y, M, and C with Bk (black) signals. Y, M, obtained in this way,
The C and Bk signals pass through a lookup table (ROM) 110 for performing color correction in this embodiment, and are converted into a correction signal Y'.
, M', C', Bk'. Then, the correction signals Y', M', C', and Bk' are sent to the print processing section 111 as shown in the conventional example, and an image is output.

<Creation of Lookup Table> The image output processing in the image copying apparatus of this embodiment has been described above. Below, the method for creating the color correction data, that is, the lookup table 110, will be explained. do.

First, the CPU 112 sends predetermined pattern signals (PG signals) Y0, M0 to the pattern generator 113 for each color of Y, M, C, and Bk.
,C0,Bk0 to generate a color patch,
The print processing unit 111 is caused to print out. When the data size of the image signal is 8-bit data, P. The G signal outputs and prints 256 gradation pattern patches of 0 to 255 for each color. FIG. 2 shows an example of a tone pattern patch printed out by the print processing section 111.

Next, this gradation pattern patch is placed on the document table glass, and subjected to the same color separation, shading correction, logarithmic conversion, and masking processing as in the image output processing described above, and is then converted into feedback signals Y, M, and C. , Bk is obtained. That is, the feedback signal is P. This corresponds to the output signal of the image output device when the G signal is input. This feedback signal is stored in memory 114.

FIG. 3 shows P. G signal Y0, M0, C0
, Bk0 and feedback signals Y, M, C, Bk. In an ideal case where there is no variation in the density characteristics of the print processing section 111, the P.I. The G signal and the feedback signal are equal. On the other hand, if the density characteristics of the print processing section 111 vary due to various factors such as the structure and external environment, a curve like ■, for example, may be formed. The G signal and the feedback signal are different signals.

Therefore, color correction is performed by P. This may be done so that the G signal and the feedback signal are the same. That is,
The lookup table 110 as color correction data is
It is created so that the curve ■ in Figure 3 is converted into a straight line ■. In other words, P
．． The CPU 112 creates a table using the G signal as data output, and sequentially writes the table into the lookup table 110. As shown in FIG. 4, this conversion is represented by a symmetrical curve (■) centered on the straight line (■) of the curve (■).

<Other Embodiments> In the above embodiments,
An example has been shown in which gradation correction is performed by referring to a lookup table stored in a ROM as a correction means. That is, the input signal is an address in the ROM, and the data at each address is represented by the curve shown by ■ in FIG. 4, and the data at the address corresponding to the signal value to be corrected is output as the corrected data, thereby making the correction. .

In the second embodiment, the above correction curve (2) is expressed as a multidimensional function Y=kn
Correction is performed by approximating +...+k2 X2 +k1 X+k0. In the following, gradation correction by calculation of an approximation function will be explained.

For example, the correction curve (■) in FIG. 4 has an input signal (horizontal axis) of X, an output signal (vertical axis) of Y, and a cubic function Y=k3
...Assume that it is expressed as (1). As mentioned above, this curve is symmetrical with respect to the straight line Y=X (■ in FIG. 3), which is the correction target, with respect to the curve ■ in FIG. 3, which is the γ characteristic of the printer. By substituting the gradation patch read signal into Y in the equation (1) and X in the equation (1), a plot on the curve (2) is plotted. (This is the curve in Figure 3 obtained when the printer input signal is set to X and the patch read signal is set to Y.)
It is equivalent to exchanging X and Y for . ) In this way, for any printer input signal Y, the patch read signal X satisfies equation (1), so for example,
64 patches of every 4th gradation are printed out and the patch density is read. In order to obtain the approximate curve function in this embodiment, it is not necessary to print out all 256 8-bit gradations as in the first embodiment.

The above 64 patches are as follows for each gradation:
Y1 =k3 X13 +k2 X12 +k1
X1 +k0 Y2 =k3 X23 +k2 X
22 +k1 X2 +k0 : :
:Y64=k3 X643
+k2 X642 +k1 X64+k0 (However,
n is the number of samples). That is,

002
6]

[Equation 1] Solve equation (2) using the least squares method (k3 , k2 , k
1, k0), the correction function equation (1) is obtained. Regarding the dimension of the correction function, since the fluctuation of the printer characteristics is approximately similar to that shown in (■) in FIG. 3, the dimension to be corrected can be determined in advance. In addition, the higher the order, that is, the higher the correction function is created, the more detailed the correction can be made. The number of dimensions is determined in advance by taking into consideration the load.

Furthermore, it is also possible to select only the necessary dimensions and create a correction function. For example, if we take the highest order up to the 5th order, then Y=k5 X5 +k4 X4 +k3 X3 +k2
If it is possible to perform a similar correction to X2 ×k1 X+k0 using Y'=k5 ×X5 +k3 X3 +k0, it is possible to reduce the calculation load and increase the processing speed. The coefficients k0 to kn thus obtained are stored in a memory, and the multiplier 501 performs the correction calculation of equation (1).

FIG. 8 shows a flowchart of gradation correction by calculation of a correction function.

(Step S81) The dimensions of the correction function are determined in advance based on the gradation characteristics of the printer. For example, if the correction function is expressed by a cubic equation, Y=k3 X3 +k2 X2 +k1 X+k0
...(1) and approximate the correction function, coefficient k0,
Determine k1, k2, k3.

(Step S82) A gradation patch is output from the printer. As the number of gradations increases (8
(Bit has 256 gradations, all gradations) Accuracy increases, but
Deriving the correction function takes time. Data may become redundant depending on the order of the approximate expression.

(Step S83) Let the printer input signal of the gradation patch be Y of the correction function (1), and let the reading signal (feedback signal) of the gradation patch be the X of the correction function (1).
Create the following simultaneous equations,

[0032]

[Equation 2] The coefficients k0, k1, k2, k3 are calculated by the least squares method.
seek. (If the printer input signal is X and the feedback signal is Y, it becomes ■ in Figure 3, and conversely, if the printer input is Y and the feedback signal is That's why.

(Step S84) The coefficients k0 to k3 are stored in memory.

(Step S85) Image data before correction (
For example, for magenta (M), cube terms and square terms of M3 and M2 are created. This can be realized using, for example, two multipliers, or may be calculated using CPU software.

(Step S86) By combining the multiplier and the adder, K3 M3 +k2 M2 +k1 M+K
0 correction. Alternatively, it may be performed by software using the CPU.

Furthermore, in the above embodiment, the number of data sizes (for 8-bit data, a patch of 256 gradations) is printed out as a gradation pattern patch, but P. It is also possible to extract the G signal discretely (for example, a pattern patch of 128 gradations every other gradation), perform interpolation processing on the missing data, and generate a correction table or correction multidimensional function. savings can be achieved.

[0037] In the above embodiment, an example was shown in which the correction table correction function is obtained separately for each color of Y, M, C, and Bk. If it can be determined that the color is not related to the color, it is sufficient to consider only a single color component as the gradation pattern patch and correction table or correction function to be output.

[0038]

According to the present invention, by appropriately correcting the complicated reproduction density characteristics of an image copying apparatus, even if the density characteristics change due to differences in structure between devices or changes in the external environment,
It is possible to provide an image correction method that can always realize good copying of images, and an image copying apparatus that applies the method.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a schematic configuration of an image copying apparatus according to the present embodiment.

FIG. 2 is a conceptual diagram for explaining the gradation pattern patch of this embodiment.

[Figure 3] P. FIG. 3 is a diagram showing the relationship between a G signal and a feedback signal.

FIG. 4 is an explanatory diagram showing the input/output relationship of color correction data.

FIG. 5 is a block diagram showing a schematic configuration of an image copying apparatus according to another embodiment.

FIG. 6 is a block diagram showing a schematic configuration of a conventional image copying apparatus.

FIG. 7 is a diagram illustrating printing characteristics of a print processing section.

FIG. 8 is a flowchart showing a procedure for tone correction using a multidimensional function.

[Explanation of symbols]

110... Lookup table (LUT), 111... Print processing section, 112... CPU, 113... Pattern generator (PG), 114... Memory M, 501... Multiplier

Claims (5)

[Claims] Claim 1: An image correction method for correcting an image in an image copying apparatus that inputs an image original and obtains a reproduced image, the method comprising: outputting an image in accordance with a predetermined gradation signal; and a step of setting a correction value based on a multidimensional functional formula so that a gradation signal created when the image is input as an image original matches the predetermined gradation signal. Image correction method. 2. An image copying apparatus that inputs an image original and obtains a reproduced image, comprising: image output means for outputting an image in accordance with a predetermined gradation signal; and inputting the outputted image as an image original. correction value setting means for setting a correction value based on a multidimensional functional formula so that a tone signal created when the tone signal is matched with the predetermined tone signal; 1. An image copying apparatus comprising: image correction means for correcting a newly input document image using correction values. 3. The predetermined gradation signal is a discrete signal, and the correction value setting means includes interpolation means for interpolating between correction values obtained from the discrete gradation signal. The image copying apparatus according to claim 2, characterized in that: 4. The correction value setting means includes coefficient setting means for setting each dimensional coefficient of a multidimensional function equation that approximates each correction value, and the image correction means corrects the image based on the multidimensional function equation. 4. The image copying apparatus according to claim 2, wherein the image copying apparatus performs the following. 5. The image correction means includes a storage means for storing a correction value using an input gradation signal as an address,
4. The image copying apparatus according to claim 2, wherein said correction value setting means stores the correction value in said storage means.