JPH03153166A - Color correction device - Google Patents

Abstract

PURPOSE:To make the capacity of an LUT(lookup table) required for color correction small in scale and to obtain a print output in 4 colors ink including a background color by applying rough color correction based on higher rank information of an input color signal, applying detailed color correction based on subsequent information of background information and applying gradation conversion to a background color signal by the conversion means (LUT). CONSTITUTION:Color signals C, M, Y converted into 3 color parallel signals are inputted to a background color detection circuit 11, from which a background color Bk(black) is detected as an 8-bit data. A selection circuit 17 uses a selection signal from a timing generating circuit 10 to select each high-order 4-bit of an input color signal and a background color signal in 8-bit and the result is outputted to a next stage LUT 12. An adder 16 adds a main color data from the LUT 12 and a correction color data generated by a multiplier 15 to form a color output. Moreover, a color output of the adder 16 and a gradation conversion output from the LUT 12 are selected by a selection circuit 18 to form a final output.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a color correction device suitable for use in a color photographic film reader or the like.

(Technical Background) As shown in Figure 5, a color photographic film reading device,
In a color image processing device such as a color scanner, an image signal read by an imaging device 1 is color corrected by a color correction device 2, and then printed by a printing device 3. The color correction device 2 adjusts the characteristics of the color separation system of the image pickup device 1 on the input side,
This is for comprehensively correcting the color according to the color characteristics of the printing device 3 (for example, printing ink) that is the output side.

(Prior Art) As the color correction device (circuit) 2, there are those that perform matrix calculations to correct colors, and those that are configured with a large-capacity memory and an LUT (look-up table).

Here, the electrification +1' circuit (method) using the LUT will be described with reference to Japanese Patent Publication No. 52-16403 titled "Title of Invention: Modification I "Image Recording Method".

This color correction circuit (method) coordinates transforms the input color signal (L B Shinyumi and CMY (No. 7) into three-dimensional grid coordinates (see Figure 6), and at that time L U T
(7) consists of only representative values, and the middle of the representative values is interpolated to greatly reduce the LUT capacity. This will be explained in detail.

Originally, a color scanner photoelectrically scans a color original image to obtain three color separation signals of R (red), G (green), and B (blue), and inputs these color separation signals into an electric city calculation circuit. Recording signals of C (cyan), M (magenta), and Y (yellow) are shown at the end of the M track. In this case, for a specific color of the color original, in order to reproduce that specific color with the printed matter (7), the combination of C, M, and Y ink amounts also takes a specific value.

That is, if the values of the three color separation signals R, G, and B are determined, C,
The amount of ink for M and Y is determined. Therefore, as shown in FIG. 6, it can be said that color correction is the coordinate conversion from the R, G, B system coordinates to the C, M, Y system coordinates.

Therefore, depending on a certain combination of R, G, and B values, C, M
, Y values corresponding to the respective R, , G, B values are stored in the memory device (LUT) in advance.

This means that it is sufficient to store the Y value in memory and read it out.

j7 mark, R, G, B values and 1. For example, if we take 200 levels of visual perception of individual color densities, the combination of three colors will produce 2003 (-8000000
) set values must be stored in a memory device, and in order to perform address i-+■ functions at high speed, core memory, 21' conductor memory, etc. must be used. Prices are expanding 1. Therefore, it is not a practical device.

Therefore, in order to reduce memory capacity, RlG,B
If the density level of each color value is decreased to 16 levels each, the combination of 3 colors will be 16'' (-4096), and you can choose an appropriate capacity (3.L or 11). If the 4th stage is too high, the output density will be 1 no. 1, and the result will be worse.
(Appropriate interpolation of the intermediate value of the floor 1. Must be done 3. For example, as shown in Fig. 6, if the output values of C, M, and Y correspond to the input values of R, G, +1, , for a certain input value (R, G,,B), the output value (C, Mo, Y) is n, and the human power value (R, G, B) n one level higher than the human power value (R, G, B) nii
17 against nil nil,
Output value (C, M, Y) is nil
When it is assumed that mouth+1 corresponds to 0+1, the interpolated values for the human power values (assumed to be R, G, and B) of these manual step values may be as follows.

Now, enter the input values (R, G, B)! :LR≦R≦
If Rn11 G ≦G≦Gni1 B ≦B≦Bn+1, then ・M)-ノ\streak ζΔM +'I n口-'Y'-1-8BXΔY ■IIn If you create an output signal, the intermediate of each stage It is possible to obtain output values that are interpolated with respect to human input values.

As is clear from the above formula, the human power values (R, G.

n Color-corrected output values corresponding to (C.B.).

Approximately, each C, M, Y output value is successively color-corrected by interpolating the intermediate part of each step output value by the uncorrected reciprocal signal n n for each C, M, Y). signal.

Here, (R-R) in the input value.

nii n (G-G), (B-B) are
Each stage n+l n n
If il n are set at equal intervals, they all become constant values, so Δ R-R-R n -C+ΔR×ΔCn n n ΔG ”” GG.

■ Δ　l　　　　　　-B-B.

■ You can calculate it using (-2).

Also, regarding (ΔG, ΔM, ΔY), In the memory device, the same 0n row as (C, M, Y) If you put these in At1nos, (R, Gn mouth B) by human signal, (80, ΔM).

11n ΔY) can also be read at the same time.

Therefore, if we take the value of C as an example, c = c + R x ΔCI.

n In order to calculate the number of people (R, G, B), From the force (C, ΔG) read from memory 12, n At the same time, calculate (ΔR) and multiply it by (ΔG), 0 shares All that is required is to add the product and (C).

(Problem to be solved by the invention) However, in normal printing, in order to correct some fogging, to make it easier to balance the gray color, and to save ink amount, -F color removal operation is required. After performing B
Printing using four-color ink, including a k (black) plate, is common. ] - It was not possible to deal with four-color ink using the method described above.

(Means for Solving the Problems) In order to solve the above problems, the present invention, as shown in FIG. - 15 color detection stage 11 for detecting a lower color signal from the input color signal; a selection stage (selection circuit 17) for switching and outputting the input color signal and the F color signal; Upper information of input color signal (-V rank 4bi
t) outputs a main color pig and difference color data between this main color data] 2. A conversion-1 stage (LUT 12) that outputs a gradation conversion output in tl I~ to the lower color signal; means (multiplier 15) for generating corrected color data from the F-order information (F-order 4 old t) of the input color signal and the difference color data;
means (adder 16) for generating a color output from the primary color data and the correction color data; and a selection stage (selection circuit 18) for switching between the color output and the step 1 conversion output to output a final output. It provides one FRA set of color values consisting of .

(Function) According to the color correction device of the present invention, rough color correction is performed using the main color data of the conversion means (LUT) based on the -position information of the input color signal, and the lower order information of the input color signal is Based on this, fine color correction is performed using the corrected color data of the converting means (LUT), and furthermore, the under color signal generated from the input color signal is gradation-converted by the converting means (LUT).

(Embodiment) An embodiment of the color correction device according to the present invention will be described in detail below with reference to the drawings.

FIG. 1 is a block diagram of the color correction device, and FIGS. 2(A) to 2(E) are timing charts.

Reference numeral 10 denotes a serial/parallel conversion circuit that converts an 8-bit three-color signal (shown in FIG. 2A) serially output from the imaging device 1 (see FIG. 5) into a parallel signal. In this embodiment, three complementary color signals of C (cyan), M (magenta), and Y (yellow) are taken as an example.

The C, M, and Y color signals converted into three color parallel signals are
The under color detection circuit 11 manually detects under color Bk (black) as 81t data.

Detected, ? = ""'T The under color detection circuit 1 has three color signals of C and M.

When it is a complementary color signal of Y, the minimum value among the three colors is used as the lower color signal and L r
When the color signals are R, G, B (7) primary color signals, the maximum value among the three colors is detected as the lower color signal.

Further, 8 from the serial/parallel conversion circuit 10
The parallel signals of three colors of bit are each 1-4 bits (
upper information) is input to the selection circuit 17, and the lower 4 bits of the valley
(lower information) is input to the parallel/serial conversion circuit 13.

14 is the (most)-position ad 1 nos of LUT 12, which will be described later.
It is a timing generation circuit that outputs 2 bits, and C, M, and Y input manually to the serial/parallel conversion circuit 10.
Divide one period of the three color signals into four equal parts, C, M, Y, Bk
It outputs a 2-bit address signal (shown in FIG. 2(C)) having timing for four color signals, and a selection signal (shown in FIG. 2(D)) which is inverted when a Bk color signal is received.

Further, the selection circuit 17 selects each of the upper 4 bits of the input color signal and y<bit according to the selection signal of the timing generation circuit IC).
The lower color signal is selected [7] and output to the next stage LUT 12 (as shown in FIG. 2(B)).

The LUT (look up table) 12 is a conversion table that converts conversion characteristics rlO,[20 to rj', and converts each manually inputted 4-bit C, Mo, Y telegraph n number (''-position information) For each of C, M, and Y, 8[)1
Each color signal data of t (-il,, input 8 bi
For the lower color signal of t, it is a memory that converts 8-bit gradation conversion data into h-. This 8-bit color signal data, as shown in the address data example shown in FIG. Stop color data ΔC' between data
, ΔM', ΔY' (at position F)
n n4 bi
It is composed of t.

That is, the conversion characteristics of this LUT]2 are shown in FIG.
- Like three-dimensional discrete grid coordinates, (C',
M', Y') 0n mouth = f'l (C, M, Y) n n n (C', M', Y')
nil mouth+l b4-1
-N(C,M,Y)nil
nii ni1Δ C' ==
-C'-C'n nii
nΔM'-M'-
M'ro Ro41 RoΔ
Y'-Y'-Y'n
The color conversion characteristic of nil n and the gradation conversion characteristic of (Bk')-r2(Bk) n are made H.

The color conversion characteristic "lO" and the gradation conversion characteristic "r20" are the color separation system characteristics of the imaging device (input end) 1 and the printing device (output side) 3.
It is preset according to the color characteristics of. Although the color conversion characteristic r10 is linear interpolation as described above, the gradation conversion characteristic r20 may be linear conversion or nonlinear conversion.

The parallel/serial conversion circuit 13 converts the F-order 4 bits (lower information) of the three-color parallel signal 81t from the manually inputted parallel signal/serial conversion circuit 10 into C, M, and Y.
Convert each 4 bits of
Output as a multiplier of 5.

On the other hand, the difference color theta of LUT12 is: i'j 15
multiplier 1s-r, differential color data 4-output of the parallel/serial conversion circuit 1() (A", \a)
Chi, the lower 41) mouth) of each color of the human power telegraph
Then, complementary+E color data is generated and input to the adder 16.

In the adder 16, the primary color data from the L iJ T 12 and the complementary iF color data obtained by the 4F in the multiplier 15 are added to produce a color output.

Furthermore, the color output of the adder 16 and the level 1 conversion output from the LUT 12 are switched by the selection circuit 18 to become the final output (as shown in FIG. 2(E)). The final output is output to the printing device 3 on the output side, and printing processing is executed.

(Effects of the Invention) As described in detail below, according to the color correction device of the present invention, the capacity of the LUT (look up double) required for color correction is small and the capacity of Bk (black) is small. ) It is possible to obtain a stamp output using four-color ink, including the undercolor of the plate.

The undercolor of '' is modified by the highly viscous nonlinear transformation A;
1a, 14, so you can get a high quality mark -81 output.
1,000 tl: Explanation No. 11': Fuu JJ 2 Me Dan Akira 1. A configuration diagram showing an example of the M-Fly fish body 1 device. Figures 2 (A) to (E) are timing charts of this device. 1. and Figure 3 are LUTs of this device.
(Look up table) - Fig. 4 is a diagram for explaining the principle of this device, Fig. 5 is a block diagram of a general image processing device, and Fig. 6 is a conventional example. FIG.

DESCRIPTION OF SYMBOLS 1... Imaging device, 2... Color correction device, 3... Printing device, 10... Serial/para 1-nor conversion circuit, 11...
Lower color detection circuit, 12... LIT (look up e-table), 13
...Rose 1/L/Serial conversion circuit, 14...Timing generation circuit, 15...Multiplier, 1 6...Adder, 17°8...Selection circuit.

Claims (1)

[Scope of Claims] A color correction device that performs predetermined color correction on an input color signal and outputs the result, comprising: undercolor detection means for detecting an undercolor signal from the input color signal; a selection means for switching and outputting the lower color signal; outputting main color data and difference color data between the main color data for upper information of the input color signal; conversion means for outputting a gradation conversion output; means for generating corrected color data from the lower-order information of the input color signal and the differential color data; and means for generating color output from the primary color data and the corrected color data. and a selection means for switching between the color output and the gradation conversion output and outputting the final output.