KR20080058893A - Color to gray conversion method - Google Patents

Abstract

A color to gray conversion method is provided to discriminate neighboring different colors through effective gray conversion when printing a color image in gray. A color to gray conversion method comprises the following steps of: preparing an original color image which will be outputted in a mono type(40); converting the same gray values into different gray values by performing gray conversion for the color image(45); performing halftoning after the gray conversion(46); and printing halftoned data in gray(47).

Description

Color to gray conversion method

1 is a flowchart illustrating a conventional scanning method.

2 is a diagram illustrating a general color image.

FIG. 3 is a diagram illustrating a gray converted image of the color image of FIG. 2.

4 is a flowchart illustrating a color to gray conversion algorithm of the present invention.

FIG. 5 is a diagram illustrating a virtual operation mask of a target pixel for labeling color images having the same gray level according to the present invention.

FIG. 6 is a diagram for describing a labeling algorithm of the present invention with a virtual operation mask of a target pixel of FIG. 5.

7 to 9 illustrate one color image, an HIS color space image, and a labeling algorithm image, respectively, to explain the effect of the present invention through the labeling algorithm of FIG. 7.

10 to 12 are diagrams showing a primary color image, a background adjustment process result image of the present invention, and an image by an existing method, to show a practical example of the background adjustment process of the present invention.

The present invention relates to a color-to-gray conversion method that processes adjacent images to distinguish when a color image is printed in black and white. Specifically, the R, G, and B values of the color image are calculated as the same black and white (gray) value. Alternatively, the present invention relates to a color-to-gray conversion method that outputs two different images to be distinguished from each other in black and white (gray) even when mapped to two different colors.

The operation of converting an existing color image to gray (gray) and outputting is performed as in Equation (1) below.

Gray (x, y) = 0.30 * R (x, y) + 0.59 * G (x, y) + 0.11 * B (x, y)

First, the R, G, B values of the pixel positions x, y in the color image composed of R, G, B are converted into a general gray calculation method according to the above equation. The constant values in this conversion are slightly different, where 0.30 for each of R, G, and B. 0.59. 0.11 is taken as an example.

An example of converting and printing a conventional color image into gray will be described with reference to FIG. 1.

According to FIG. 1, the input color image 10 is converted by a color-to-gray conversion method such as Equation (1), and the calculated gray (x, y) value is represented by the gray gamma table 14. The gray value is converted into gray values suitable for the printer engine through the gray half toning table 16, and then transmitted to the printer and printed (18) according to the gray dither half toning table.

Although there are various gray conversion formulas for R, G, and B color values, as shown in Equation (1), R, G, B values are different, but as a result, they may have the same gray value after gray conversion. Will occur.

As shown in Figs. 2 and 3, if two color images (Fig. 2) are immediately adjacent to each other after the gray conversion of the R, G, and B values, they are calculated as the same gray values, so they are mono. In the printed output, there is no difference in the image, so that the distinction is difficult (Fig. 3). Also, even if an algorithm is used to output two different colors that are converted to the same gray color, the output of the two grays that the user sees may look the same or be gamma-gray depending on the characteristics of the printer. If you have a Gamma Table, only the calculated gray you want will be different, and the half-toned values going to the printer will be meaningless, and you will not be able to distinguish them in the final output.

As other prior arts, Korean Patent Laid-Open Publication No. 2002-057825 converts color data constituting original data consisting of image data having a plurality of colors and the like into luminance data, and converts colors that can be printed by a printer into luminance levels of respective pixels. Disclosed is an image processing apparatus and method capable of creating print data based on a luminance level of an original data image by allocating to each pixel.

In another Korean Patent Laid-Open Publication No. 2000-066828, halftone processing is performed in an error diffusion method in which a gray scale error value between a pixel value of a continuous gray scale image and a thresholded binary pixel value is diffused to adjacent pixels according to a predetermined weight to perform binarization. After subtracting the gradation error value of the neighboring pixel adjacent to the pixel value of interest to be performed, a random value is calculated and then added to the first correction pixel value to calculate the second correction pixel value to perform an error diffusion process. Is disclosed.

On the other hand, Japanese Laid-Open Patent Publication No. 2005-246665 discloses text based on an identification means for extracting a text portion from an image portion and a text portion mixed in the printed image data and a density recognition means and a density recognition means for the text portion extracted by the identification means. Disclosed is an image forming apparatus having a density converting means for making the density of a portion the maximum density or the minimum density, and a printing means capable of obtaining achromatic color by two or more inks and printing them in gray scale.

In addition, US Patent No. 5,737, 453 discloses an apparatus for processing error diffusion halftoning by providing a different error dispersion matrix for each gray level value and selecting an error dispersion matrix according to the gray level value of the target pixel of the original image.

There was no teaching on the case where they were indistinguishable in the output when the gray conversion values of the color images were the same.

According to the present invention, in order to achieve the above-described problems, when a color image is output in gray, when two images having completely different colors are adjacent to each other, the gray is more effectively converted so that the color-to-gray conversion values of the two color images are not the same. It is an object of the present invention to provide a color-to-gray conversion method that converts a value to another value so that two colors can be distinguished even when outputted in mono (mono) color.

In order to achieve the above technical problem, the present invention, in the color-to-gray conversion method, the step of preparing the original color image to be largely output in mono, by converting the gray for the color image to change the same gray value to different gray values And a halftoning step after the gray conversion, and outputting the halftoned data in gray.

Specifically, the gray value conversion and change step uses the color value in the HIS color (Color, Brightness, Saturation: Hue, Intensity, Saturation) color space as the input value of the edge operator. step; Distinguishing objects having a series of identical gray values through a labeling operator using the edge values; And determining whether to adjust the gray value by calculating the distances of the separated locations.

Preferably, the gray value is not changed if the distance of the pixel positions of the labeled values to the gray value is greater than or equal to the specific value.

Hereinafter, the operation of the present invention will be described in detail with reference to the accompanying drawings.

When the printer outputs, the user wants to output a color image as a mono image. When the output signal is pressed, R, G, and B data of the entire color image are transferred to the memory buffer. Convert to HIS color space for the entire image. The conversion formula to the HIS color space can be applied in various ways, but is generally as shown in Equation (2) below.

I = (R + G + B) / 3

S = 1-3 * min (R, G, B)

H = COS ^-1 [{(RG)-(RB) {/ {2 * sqrt ((RG) ^ 2 + (RB) (GB))}]

Here, I is brightness, S is saturation, H is hue, and satisfies 0 ≦ R, G, B ≦ 1.0. min is the function that takes the minimum value, and squart is the function of the square root.

By using the above conversion formula, HSI (x, y) for the pixel values R, G, B (x, y) of the corresponding position x, y in the entire image can be obtained. In this equation, since the colors are different and luminance, i.e., I must find the same value, we can find the x, y set of candidate regions as below.In the above equation, luminance I is the same, but color H (x, y) is different Candidate areas can be found.

In an 8-bit processing system expressing a gray level from 0 to 255 as a set that satisfies this expression, a set that satisfies the expression for each gray level may be represented by Equation (3) below.

HI [grayLevel] = {H0 (x0, y0), H1 (x1, y1), H2 (x2, y2) ....}

In other words, the meaning of this expression is the same gray level, but there are pixels of x0, y0, pixels of x1, y1, pixels of x2, y2, etc., where the color is different. Mean H0, H1, H2, ... respectively.

Then, you can see where the colors are different but the gray level values are the same. However, you have to worry about whether you should do this specially. If the colors of the two colors are different and the gray values are the same, but if the distances of these two colors are far apart from each other when viewed in the whole image, the gray values from the other color images are not distinguished in the middle. This is because the two colors can be distinguished from each other.

What distinguishes the two colors before the gray conversion is the outline of the two colors, which can be seen as the outline of the color values. Thus, an outline value for Hue can be obtained from the entire HIS color space. In this case, the general contour detection operator may be used.

According to FIG. 4, after the color image is prepared (40), an edge value for the entire color value is obtained (41) using the contour detection operator as described above, and the pixels exceeding a certain threshold value or more for this edge value. When the value of x, y is called E (x, y), the value of E (x, y) is expressed by the following equation (4).

E (x, y) = 1 IF EdgeValue (x, y)> T (threshold)

E (x, y) = 0 else

Next, labeling operations of colors having the same gray level value in each gray color area are performed, and the index to the target pixel position P (x, y) for operation is P4 and the virtual The operational mask is shown in FIG.

In the present invention, the algorithm for labeling is calculated by scanning in four directions of up, down, left and right, that is, the entire image, as shown in FIG. 6.

That is, Top-> Bottom and Left → Right (TBLR), Top-> Bottom and Right → Left (TBRL), Bottom-> Top and Left-> Right (BTLR), Bottom-> Top and Right-> Left. In addition, the value obtained here is represented by a label [gray].

The labeling algorithm is as follows.

(Labeling algorithm)

For GrayLevel = 0 To 255

LabelIndex = 0

If H (x, y) is one of elements of HI [GrayLevel]

If E (x, y)! = 0 Then

If TBLR calculation

P4 = P1 * P3 * (P0 + P3)

If TBRL calculation

P4 = P1 * P5 * (P2 + P5)

If BTLR calculation

P4 = P7 * P3 * (P6 + P3)

If BTRL calculation

P4 = P7 * P5 * (P8 + P5)

If P4> = 1 Then P4 = 1

Copy of E (x, y) = P4

If P4 = 1 for all scan direction Then

Next GrayLevel

Through the labeling operation, an effect as shown in FIGS. 7 to 9 can be obtained. 7 is the original color image, where the two adjacent circles above are different in color from H0 (blue) and H1 (purple), but the gray values are the same as G0, and the two adjacent circles below are in color H2 (light blue). ) And H3 (yellow), but because the gray value is the same as G3, the gray value in the adjacent area is the same as shown in Figure 8 when converted to HIS color space image was not distinguished. However, as shown in Fig. 9, in the image conversion by the labeling algorithm, the change of the gray value in the adjacent area is different between a and b.

Referring back to FIG. 4, finally, the distance of pixels with different labeling at the same gray level is calculated (43). Then it is checked if a change in gray value is needed (44). If the distance of the pixel positions of the labeled values for a certain gray value is more than a certain value, then the gray value is not changed.

That is, such an algorithm is constructed as follows.

(Gray value change algorithm)

For all element of each Label [Gray] a vs. Label [Gray] b

If Distance between a (x, y) and b (x, y) <T

change Gray of one of Label [Gray]

Else

Skip

With the above algorithm, we can convert the gray values of adjacent gray levels. This is illustrated in FIGS. 10 to 12.

As shown in Figures 7-9, Figure 10 is the original color image, where the two adjacent circles above have different colors in H0 (blue) and H1 (purple) but the gray values are the same in G0, The two adjacent circles differ in color from H2 (light blue) and H3 (yellow), but the gray value is the same as G3. Therefore, the gray values in the adjacent areas are the same as shown in FIG. The area of the circle was not distinguished. However, as shown in Fig. 11, in the image conversion by the labeling algorithm of the present invention and the distance to the pixel position, the difference can be displayed in the change of the gray value in the adjacent region. In other words, the above adjacent circles are distinguished by different gray values, and the below adjacent circles are applied to the same gray value so that they are not distinguished.

As described above, according to the present invention, when printing a color image in gray, it is possible to output the color of another adjacent color to be distinguished through an effective gray conversion.

Claims (3)

In the color to gray conversion method, Preparing an original color image to be output in mono; Gray converting the color image to change the same gray value to different gray values; Halftoning after gray conversion; And And outputting the half-toned data in gray. The method of claim 1, wherein the converting and changing the gray value is performed. Using a color value in an HIS (Hue, Intensity, Saturation) color space as an input value of an edge operator when converting gray; Distinguishing objects having a series of identical gray values through a labeling operator using the edge values; And And determining whether to adjust the gray value by calculating the distance of the separated positions. 3. The method of claim 2, wherein the gray value is not changed if the distance of pixel positions of the labeled values to the gray value is greater than or equal to a specific value.

Images