JP2547746B2 - Color image processing method - Google Patents

Description

TECHNICAL FIELD The present invention relates to a color image processing method for performing color correction.

[Prior art]

FIG. 2 shows a conventional general configuration example of such a color copying apparatus. The color original 11 is illuminated by an illumination light source 12, is imaged on a photoelectric converter 13 by an imaging lens, and is R (red), G (green), B (blue) 3 by a color separation filter of three colors.
Converted to primary color analog electrical signal. The R, G, B signals are A
A / D converter 14 converts it to a digital signal, and a complementary color converter 15
And subtractive mixed primary color signals Y (yellow), M (magenta), C
It is converted to (cyan), further logarithmically converted by the logarithmic converter 16, color-corrected by the masking circuit 17, and sent to the printer 18. Here, the color correction calculation of the masking circuit 17 cancels the influence of unnecessary absorption of the dye used in the printer, so that the input signals Y _in , M _in , and C _in are expressed by the following equation (1).
Output signals to the printer, Y _out , M _out , and C _out .

Nij in the equation (1) is generally called a masking coefficient.

The printer 18 outputs a multi-valued image by using a technique such as dither depending on Yout, Mout, and Cout in the equation (1).

However, the masking coefficient largely depends on the color separation characteristics of the photoelectric converter 13, the output characteristics of the printer 18, etc., and for example, the coefficient for obtaining the optimum color reproducibility in the configuration of FIG. 2 is determined. Even if you do so, if the input system changes over time, or if you consider the connection with other input systems and printers, you must change the masking coefficient as appropriate, which is impossible in practice. Met.

〔Purpose〕

An object of the present invention is to optimize matching regarding color reproduction in a system including a reading unit and an output unit so that good color reproduction can always be performed.

〔Example〕

Examples of the present invention will be described in detail below. First, prepare sample color batches of multiple colors, and
Let chromaticity coordinates L ^* , a ^* , b ^* in the b color system be L _i , A _i , B _i . It is desirable to select the sample color so that its chromaticity is evenly distributed in the uniform color space.

Next, the output characteristics of the printer are obtained by the following procedure.

Step 1: Given arbitrary input signals U _j , V _j , W _j to the printer, obtain chromaticity L _j ^out , A _j ^out , B _j ^out of the output color at that time. Input signals u, v, w are usually Y (yellow), M (magenta), C
The (cyan) signal is used, but here we will write u, v, w.

Step 2: ▲ L ^out _j ▼, ▲ A ^out _j ▼, ▲ B ^out _j ▼ u _j , v _j ,
Since it is a function of w _j , this inverse transformation is obtained. That is, The coefficient m is determined by calculation so that the following equation is satisfied by a large number of j at the same time.

Step 3: The printer input signal values u _i ′, v _i ′, w _i ′ corresponding to the chromaticities L _i , A _i , B _i of the sample color patch are back-calculated by the equation (2). That is, From equation (3), u _i ′, v _i ′, w _i ′ are calculated for the chromaticities of all color samples, and these are calculated in RO in the printer.
Write it in M etc.

The above is the procedure for obtaining and storing the printer characteristics.
Next, the masking coefficient is optimized according to the characteristics of the input system. This structure is shown in FIG. Reference numeral 21 is a color original containing multicolored samples, 22 is a signal processing unit in which two or more of FIGS. 13 to 16 are put together, 23 is a frame memory for temporarily storing output signals of the input original, and 24 is specified from the frame memory 23. A pixel extraction device for selecting and reading out the signal value of a pixel, 25 a micro computer for calculating masking coefficients by calculation, 26 a masking RAM for writing the coefficients determined by the micro computer 25, 27 obtained by the color (3) above The printer characteristic ROM that stores the obtained u _i ′, v _i ′, w _i ′ for all color samples.

 The procedure for determining the masking coefficient is shown below.

Step 1 ': The pixel corresponding to the i-th color sample is extracted from the image stored in the frame memory 23 by the pixel extracting device 24, and the output signal values x _i , y _i , z _i are read.

Step 2 ′: If x _i , y _i , z _i match with u _i ′, v _i ′, w _i ′ obtained by the formula (4) after masking, the optimum color reproduction for the i-th sample is obtained. available.

That is, the masking coefficient is n _kl Should be valid for all i. You can obtain n _kl by calculation.

Step 3 ': The calculation result of the microcomputer 25 is written in the masking RAM 26. After that, the output from 22 is
While referring to the RAM 26 in the masking processing unit 17, the input system,
An optimized masking operation will be executed in consideration of the characteristics of both printer output systems.

In FIG. 1, 23, 24, and 25 do not have to be built in the input device at all times, and may be externally connected only when writing 26 RAMs. Further, in the case of the present embodiment, the frame memory 23 does not require high pixel density, and therefore an extremely low capacity is sufficient. However, if a frame memory is already built in the input device, this is referred to. It doesn't matter.

As described above, since it is possible to easily optimize the input system with the printer output system based on the color sample specified in advance, for example, when the color separation filter of the input system is discolored due to a change with time, etc. Also, by performing optimization each time, the initial color reproducibility can be secured. There are various types of printers such as electrostatic photography, ink jet, thermal transfer, etc., but even for printers that originally have different color reproducibility and color mixing systems, the compatibility can be established by referring to the printer ROM. I can do things.

Further, for example, when trying to copy a color manuscript that is discolored to a bluish color, if a blue filter is placed on the color sample 21 in FIG. On the other hand, it is also possible to obtain a copied image from which blue is removed.

〔effect〕

As described above, according to the present invention, even if the input characteristic of the reading unit or the output characteristic of the output unit changes, the matching regarding color reproduction in the system including the reading unit and the output unit is optimized, and good color reproduction is always performed. Can be made possible.

Further, since the input characteristic of the reading unit and the output characteristic of the output unit are independently obtained and the color correction parameter is generated based on the independently obtained input characteristic and output characteristic, for example, in the system, the reading unit and the output unit Even if the combination of is changed, it can be flexibly dealt with, and good color reproduction can be performed by the system.

[Brief description of drawings]

FIG. 1 is a control block diagram of the color copying apparatus of this embodiment,
FIG. 2 is a control block diagram of a conventional color copying apparatus.

Claims (1)

(57) [Claims] 1. Color correction in a system including a reading unit for reading a target image to obtain color image data representing the target image and an output unit for outputting a visible image based on the color image data is performed based on a color correction parameter. A color image processing method, comprising: reading output characteristics of the output unit generated based on a color patch output by the output unit based on desired color image data, and reading a color reference image by the reading unit to obtain the color reference image. A color image processing method, comprising: reading an input characteristic of the reading unit generated based on the obtained reference color image data, and generating the color correction parameter based on the output characteristic and the input characteristic.