JPH06324654A - Binarizing method for color image signal - Google Patents

Abstract

PURPOSE:To provide the binarizing method for the color image signal which can output a binarized signal corresponding to the color components of the color image signal to be outputted by using the color components of an inputted color image signal. CONSTITUTION:By this binarizing method, the binarized signal is generated from the basic color signal of the input image signal by adding a color signal corresponding to the basic color signal and an achromatic signal; and color areas 1000 and 1001 of the achromatic color signal and the color area of the composite color component of the color signal corresponding to the basic color signal and the achromatic color signal are previously set, the color signal is determined by giving selection conditions of the color areas 1000 and 1001 of the achromatic color signal priority to selection conditions of other color areas, and the color signal corresponding to the achromatic color signal and basic color signal is binarized.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a binarization method for a color image signal, which inputs a color image signal and generates color signals having different color gamuts for output / display.

[0002]

2. Description of the Related Art A conventional binarization method for color image signals is as follows.
It was done as follows. (1) Three color components of the input color image signal (for example, R
The GB) signal is binarized for each color component signal. Then, when the signal amounts of the respective color components match, an achromatic color signal is generated. (2) Three color components of the input color image signal (for example, R
From the GB) signal, an achromatic signal is generated by setting a color signal component common to each color component signal as an achromatic color. Then, a new three-color component signal is generated by subtracting the amount of the achromatic color signal from each of the original three-color component signals. Each of the four color component signals including the achromatic color thus obtained is binarized.

[0003]

The method (1) described above can be realized by relatively simple hardware, but it has a drawback that the image quality is not always good because the achromatic component is simply generated.

Further, the method shown in (2) requires four independent binarization circuits for binarizing the image signal,
There was a problem implementing it in hardware. For example, if the error diffusion method is applied to the method (2), four kinds of binarization circuits by error diffusion are required, which is a heavy load on the circuit configuration.

When the error diffusion method is applied to the method (1), there is no problem if the achromatic color component is the same as the basic three-color signal composite signal, but in general, the independent achromatic color signal and three-color signal are independent. Have a different color component from the achromatic signal obtained by synthesizing. Therefore, in the case of simply binarizing the three-color component signals, if the values of the three-color component signals coincide with each other, if they are forcibly replaced with a specific achromatic color signal, the apparent colors will differ, and the reproduced image There was a problem such as deterioration of image quality.

The present invention has been made in view of the above-mentioned conventional example, and a binary image of a color image signal capable of outputting a binarized signal corresponding to a color component of a color image signal to be output from a color component of an input color image signal. The purpose is to provide a method of conversion.

[0007]

[Means for Solving the Problems] and

In order to achieve the above object, the method for binarizing a color image signal according to the present invention comprises the following steps. That is, a binarization method of a color image signal for generating a binarized signal by adding a color signal corresponding to the basic color signal and an achromatic color signal from the basic color signal of the input image signal, Of the color gamut, the color gamut of the color signal corresponding to the basic color signal and the color gamut of the composite color component of the achromatic color signal is preset, and the selection condition of the color gamut of the achromatic color signal is a selection condition of another color gamut. The color signal is determined with higher priority, and the achromatic color signal and the color signal corresponding to the basic color signal are binarized.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT A preferred embodiment of the present invention will now be described in detail with reference to the accompanying drawings. Before describing the present embodiment, the pyramid error diffusion method will be described.

<Description of Pyramid Error Diffusion Method> FIG. 4 is a block diagram showing the configuration of a circuit for performing quantization by the error diffusion method.

In the figure, 1 is an A / D converter which converts the input analog image data into digital data for each color component (RGB). 2 is a gamma conversion unit for performing gamma (γ) correction, 3 is a masking unit for performing color correction,
Reference numeral 4 is a quantizer that binarizes and quantizes digital data.

The operation of the circuit having the above configuration will be described step by step. The input analog RGB color signal is A /
In the D converter 1, each color is converted into an 8-bit digital RGB signal. This RGB digital signal is input to the gamma conversion unit 2, and gamma correction is performed so that the gamma characteristic of the input RGB color signal matches the gamma characteristic of the output device (such as CRT). Here, for example, when it is desired to convert an input color signal having a gamma characteristic of 0.45 power obtained from a television camera or the like into a signal having a linear gamma characteristic, 2. This is equivalent to performing squared gamma correction. Next, in the masking section 3, color correction of RGB color signals is performed. This is represented by a 3 × 3 linear conversion equation as shown in the following (Equation 1).

[0012]

[Number 1] _{R = m 11 × R + m} 12 × G + m 13 × B G = m 21 × R + m 22 × G + m 23 × B R = m 31 × R + m 32 × G + m 33 XB Finally, the quantization unit 4 quantizes the 8-bit RGB color signals of each color into R, G, B, W color signals of 1-bit for each color. The quantizer 4 will be described in detail with reference to the drawings.

Hereinafter, a case where each color signal is binarized in the quantizer 4 will be described as an example. Unless otherwise specified, basically the same applies to cases of three or more values. The output device used in this description is not a printer that prints using C (cyan) M (magenta) Y (yellow) K (black) type color ink, but R (red) and G (green). , B (blue), W
Use a color monitor with (white) as the primary color.

FIG. 5 is a diagram showing a color reproduction space of a color monitor having the above-mentioned RGBW as primary colors. Reference numeral 500 shows a color reproduction space by the RGB system, and 501 shows a color reproduction space by the RGBW.

A characteristic of the color reproduction space of the RGB color monitor is that eight colors (black, black, etc.) of each combination of RGB are characteristic.
R, G, B, RG, RB, GB, RGB) color space and 8 colors (W, RW, G) in which W is further combined.
W, BW, RGW, RBW, GBW, RGBW), and two color spaces 500 and 501 are formed to overlap each other. That is, the output device that uses the three primary colors of RGB as the primary colors can express eight colors, while the RGBW
There are 16 colors that can be expressed by an output device that uses four primary colors.

Therefore, in the present embodiment, the input image data of the RGB three colors is shown without generating the auxiliary color signal W from the input image data of the RGB three colors before the quantization processing by the quantizing unit 4. The nearest color in the output 16 colors is calculated with respect to the color. Then, the error generated between the input image data of RGB three colors and the nearest output color is error-diffused to the points near the processing pixel.

Here, in order to binarize the auxiliary color signal W without generating it in advance, it is necessary to use 1 in the RGBW color space.
The six colors and the positional relationship indicated by the input image data of the RGB three colors must be associated in advance. This correspondence is shown in FIG. In FIG. 8, each of the input RGB data is represented by 8 bits.

In the case of this embodiment, the ratio of the luminance value when all the RGBWs are turned on to the luminance value when only the three RGB colors are turned on is 255: 153. Similarly, the luminance ratio of RGBW and W is the same. Since it was 255: 102, it is simply associated. Of course, as shown below, it is more preferable to make color correspondence. That is, (1) each of the four colors of RGBW is measured and the individual tristimulus values XY
Find Z. (2) The tristimulus values XYZ of the output 16 colors due to the combination of RGBW are obtained from the following equation (Equation 2).

[0019]

## EQU2 ## X = R × rX + G × gX + B × bX
+ W × wX Y = R × rY + G × gY + B × bY + W ×
wY Z = R × rZ + G × gZ + B × bZ + W ×
wZ where rX, rY, and rZ are tristimulus values of R, gX, gY,
gZ is the tristimulus value of G, bX, bY, and bZ are the tristimulus values of B, and wX, wY, and wZ are the tristimulus values of W.
(3) From the tristimulus values XYZ of the output 16 colors, for example, NTS
RGB of C system is calculated by the following equation (Equation 3).

[0020]

R = 1.9106X-0.5326Y
-0.2883Z G = -0.9843X + 1.9984Y-0.
0283Z B = 0.0584X-0.1185Y + 0.
8985Z By the above operation, the correspondence between the input RGB data and the output color can be easily and chromatically obtained. As a result, RG
From the B three-color input image data, the auxiliary color signal W
It is possible to obtain the nearest neighbor color for outputting 16 colors with respect to the color indicated by the input image data of three colors of RGB without generating the B before the binarization processing. Further, by diffusing the error generated between the closest color and the input image data to the neighboring points, high-quality binarization processing without color unevenness is possible. Of course, the same applies to the case where the output colors of the CMYK four-color data can be output when the CMY three-color data is binarized.

A detailed description will be given below with reference to the drawings.

FIG. 6 is a block diagram showing a schematic configuration of the quantizer 4.

The input data f _mn indicates the density data of the pixel at the coordinate (m, n) point input from the input unit 101. An adder 102 adds the accumulated error amount x _n from the line buffer memory 112 and the input data f _mn . The line buffer memory 112 stores data that is weighted and cumulatively added by referring to the error diffusion table 111. Thus, g _mn (g _mn = x _n + f _mn ) obtained by adding the cumulative error x _n to the density data f _mn is 2
It is input to the digitization circuit 103. The binarization circuit 103 obtains the binary data D _mn of the output device color closest to the input data g _mn and the corresponding data B _mn, and outputs the binary data D _mn to the output unit 104 to obtain the data. B _mn is output to the error calculation unit 107.

The error calculation unit 107 calculates (B _mn −g _mn = e
_The error en is _calculated from ( _n ) and the result is output to the error diffusion table 111. Using the error diffusion matrix in the diffusion table 111 stored in the line buffer memory 112 performs a predetermined weighting to the error e _n. For example,
Assuming that the error so far is stored as shown in the line buffer memory 112 of FIG. 6, the error when processing the pixel at the position of x _{n + 1} is newly _expressed as x _{n + 1} <−x _{n + 1} + (2/7) x e _n x _{n + 2} <-x _{n + 2} + (1/7) x e _n x'n _-2 <-x ' _n-2 + (1/7) x a _{e n x 'n-1 <} -x' n-1 + (1/7) × e n x 'n <-x' n + (2/7) × e n. When the scanning of the original image data for one line is completed, the first line of the line buffer memory 112 contains the data of the second line and the second line contains "0". Binarization processing is performed by repeating such processing. The output unit 104 then outputs the binarized data D
_The dot is turned on / off according to the value "1" or "0" of _mn , and the quantized data is displayed / output.

FIG. 7 is a flowchart showing a binarization (quantization) process for obtaining the nearest color when the output color is 16 colors with respect to the color indicated by the RGB input image data.

First, in step S1, an initial value k = is set to a parameter M used when obtaining the counter k and the nearest color.
After substituting 0, M = 9999, the distance L between the input data RGB and the data of the kth output color is obtained from the following equation in step S2.

[0027]

Equation 4] _{L = (R-R 0)} 2 + (G-G 0) 2 + (B-B 0) 2 where, R, G, B: RGB data R ₀ inputted, G _0, B _0: Output Color RGB Data In step S2, the distance L between the input RGB data and the kth grid point data is obtained, and it is determined whether or not the distance L is smaller than the minimum value M. If L <M, the process proceeds to step S4, the value of the distance L is set to the minimum value M, and the value of the counter k at this time is stored in the variable kk. Next, in step S5, the counter k is incremented by 1 to see if the value of the counter k is equal to or larger than the predetermined number n.
When the value of the counter k is smaller than "n", step S2
Then, the above process is repeated.

In this way, steps S2 to S6 are repeated by a predetermined number n, and when the value of the counter k becomes equal to n or becomes larger than n, the process proceeds to step S7.
The grid point at which the minimum distance L was obtained is determined based on the value of the variable kk. Then, in step 7, the output color data table (binary data) br, bg, bb, bw
And the RGB input data tables tr, t corresponding to them
By accessing the kk addresses of g, tb, and tw, the respective binary data D _mn _r, D _mn _g, D _mn _ are accessed.
b, it is possible to obtain the D _mn _w, RGB input data B _mn _r corresponding to each of _{these, B mn _g, B mn _b} , and B _mn _w.

<Description of Embodiments> FIG. 1 is a conceptual diagram of the above-described pyramid ED. Here, the color coordinate system is basically described based on the color coordinate system of the output device. Further, the primary color points of the output device are R1, G1, B1, W, and here, it is assumed that the additive color mixture is established based on the display system of the output device. Since the output device is basically binary recording or binary display, the output basic color point is R1,
G1, B1, W, WR1, WG1, WB1, WR1G1
B1, R1G1, G1B1, B1R1, WR1G1, W
There are 16 colors in total including G1B1, WB1R1, WR1G1B1, and Bk.

Now, output achromatic color components W, WR1G1B
The color gamut ranges for the two points 1 are set as follows.

If the color gamut of Wr ₁₁ ≦ R ≦ Wr ₁₂ Wg ₁₁ ≦ G ≦ Wg ₁₂ Wb ₁₁ ≦ B ≦ Wb _12, then W = 1.

If the color gamut Wr ₂₁ ≦ R ≦ Wr ₂₂ Wg ₂₁ ≦ G ≦ Wg ₂₂ Wb ₂₁ ≦ B ≦ Wb _{22 of} WR1G1B1, W = 1, R1 = 1, G1 = 1, B1 = 1.

The color gamut within the set values is preferentially W, WR1
The achromatic signal of G1B1 is selected. In the case of FIG. 1, the color gamut indicated by 1000 is W = 1,
The area indicated by 1001 is an area of W = R1 = G1 = B1 = 1.

In the color gamut other than the above, color signals of three colors to be output are selected under the following conditions.

Color Gamut of R1 0 <R ≦ R1 R1 = 1 G1 Color Gamut 0 <G ≦ G1 G1 = 1 B1 Color Gamut 0 <B ≦ B1 B1 = 1 WR1 Color Gamut R1 ≤ R ≤ If WR1, R1 =
1, W = 1 Color gamut of WR1 G1 ≤ G ≤ If WG1, G1 =
1, W = 1 Color gamut of WR1 B1 ≤ B ≤ If WB1, B1 =
1, W = 1 Under other conditions, W = R1 = G1 = B1 = 0.

The determination of the above conditions can be performed quickly and easily with the circuit configuration shown in FIG.

The circuit shown in FIG. 2 shows the configuration of a circuit for calculating the color gamut of W, and 230 is a circuit applied to the R (red) component. Each of 201 and 203 is a threshold value of the R component, and Wr ₁₁ and Wr ₁₂ are given respectively. Reference numeral 204 denotes a complement circuit. The output of the complement circuit 204 becomes -R, (Wr ₁₁ -R) is calculated by the adder 205, and the adder 206
Then, (Wr ₁₂ −R) is calculated. These adders 205,
Since the difference calculation result by 206 is output as the sign bits 220 and 221 of the adders 205 and 206, these sign bits 220 and 221 are judged by the AND circuit 207, and the color data of the R component is Wr ₁₁ ≦ R ≦ It can be determined whether or not the condition of Wr ₁₂ is satisfied. 209 is the circuit 230
Is a circuit in which G is applied to the G (green) component, and 210 is a circuit in which the circuit 230 is applied to the B (blue) component. In this way, the output of the AND circuit 208 becomes high level when all of these judgments are true, and this area is shown in FIG.
Area (W = 1).

Although FIG. 2 shows a circuit configuration for generating the W component signal, similarly, WR1G1B1, R1, and
It can be seen that the area for generating each component signal of G1, B1, etc. can be determined quickly and easily.

FIG. 3 is a diagram for explaining a method of obtaining a new pyramid ED and binarizing it based on the above-described concept.

In FIG. 3, 400 is the input RGB
The three-color component signals represent red (R), green (G), and blue (B) color image signals transmitted from a host computer or the like. Here, for simplification of description, it is assumed that the color signal range of the input color image signal is normalized so as to match the range in which the color of the output device can be expressed. As a result, if the color gamut of the input image and the color gamut of the output device are different, the apparent colors are different, but the colors can be identified. A conversion circuit 401 realizes such a function.
When the output color gamut and the input color gamut are different from each other, it is a circuit portion that adjusts them so that they match each other. Therefore, in the following description, the output of the conversion circuit 401 is treated as R, G, B signals.
An adder 402 adds the data sent from the memory 413 of the error diffusion part. The output of the adder 402 is input to the color gamut determination circuit 403 on the one hand and the subtracter 410 on the other hand.

The color gamut discrimination circuit 403 is composed of the circuit as shown in FIG. Gamut discrimination circuit 403
Output four color component signals (R1, G1, B1,
The W) combination 404 is output, while the color binary signal is directly generated through the signal line 409. On the other hand, the output 404 is input to 405, and is composed of a combination of four color component signals (R1, G1, B1, W) 16
A set of data in the color component table 406 (see FIG. 8) is selected according to the values of the four color component signals. The data thus selected is represented by (R ₀ , G ₀ , B
₀ ). This value is input to the subtractor 410, and the difference (ΔR, ΔG, ΔB) from the input data (RGB): ΔR = R
−R ₀ , ΔG = G−G ₀ , ΔB = B−B ₀ ) is calculated. Reference numeral 412 is an error diffusion table that has been conventionally applied. Reference numeral 413 is a memory for storing the error diffused by the error diffusion table 412. Memory 4
A part of the diffusion error stored in 13 is fed back to the adder 4
02 is added and the original image data is saved. Reference numeral 408 represents a binarized signal state of the color image output to the signal line 409.

As is clear from the above description, the gamut determination circuit 403 determines the closest value between the input three-color component signal and the reference three-color signal generated from the four-color component signal of the output device. By implementing the above, it becomes possible to process a combination of 16 colors for one set of input data by a program much faster than the conventional method of obtaining the nearest color from the sum of squared differences of each color component. Therefore, the speed of recording or displaying a color image can be increased.

In the present embodiment, the case of the additive color mixture based on the display color system has been described. However, even in the case of subtractive color mixture using ink or the like, the speed can be increased by the same method as in the present embodiment. It's obvious.

In this embodiment, the RGB color system has been described, but the same speedup can be achieved by using other non-linear color system such as CIE LA ^* B ^* system.

[0045]

As described above, according to the present invention, there is an effect that a binarized signal corresponding to the color component of the color image signal to be output can be output from the color component of the input color image signal.

[Brief description of drawings]

FIG. 1 is a conceptual diagram of pyramid error diffusion (ED).

FIG. 2 is a block diagram showing a configuration of a circuit for calculating a W color gamut.

FIG. 3 is a diagram for explaining a method of obtaining a new pyramid ED and binarizing it.

FIG. 4 is a block diagram showing a configuration of a circuit that performs quantization by an error diffusion method.

FIG. 5 is a diagram showing a color reproduction space of a color monitor having RGBW as primary colors.

FIG. 6 is a block diagram showing a schematic configuration of a quantization unit.

FIG. 7: Binarization (quantization) for obtaining the nearest color when the output color is 16 colors with respect to the color indicated by RGB input image data
It is a flowchart which shows a process.

FIG. 8 is a diagram showing 16 colors in the RGBW color space and the positional relationship indicated by RGB input data in association with each other.

[Explanation of symbols]

 204 Complement circuit 205, 206, 402 Adder 400 Input RGB 401 Conversion circuit 403 Color gamut determination circuit 410 Subtractor 412 Error diffusion table 413 Memory

Claims (1)

[Claims] 1. A binarization method of a color image signal for generating a binarized signal in which a color signal corresponding to the basic color signal and an achromatic color signal are added from a basic color signal of an input image signal. The color gamut of the achromatic color signal, the color gamut of the color signal corresponding to the basic color signal and the composite color component of the achromatic color signal are preset, and the selection condition of the color gamut of the achromatic color signal is set to another color gamut. 2. The method for binarizing a color image signal, wherein the color signal is determined with priority over the selection condition of step 2 and the achromatic color signal and the color signal corresponding to the basic color signal are binarized.