JPH05334410A - Color image display method - Google Patents

Abstract

PURPOSE:To efficiently reproduce picture quality corresponding to a color image consisting of a group of picture elements by arranging the color image in an RGB space based upon RGB data. CONSTITUTION:In a color image consisting of a group of picture elements, the picture elements are arranged on an RGB space consisting of a three- dimensional space using respective colors R, G, B as respective rectangular coordinate axes based upon RGB data. The maximum gradation levels of R, G and B components are defined as Rm, Gm and Bm and their minimum gradation levels are defined as Ro, Go and Bo. Out of all picture elements in the color image including a group of picture elements forming an approximately elliptic shape dispersed around the periphery of an achromatic color line OF obtained from the position O of coordinates (Ro, Go, Bo) and the position F of coordiantes (Rm, Gm, Bm) in the RGB space, only the picture elements forming the elliptic shape are extracted as colors necessary for reproduction. Since automatic processing based upon a computer is simple and rapid processing is available almost real-time display can be attained.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of previously extracting color image data required for reproduction from an input color image.

[0002]

2. Description of the Related Art Generally, an output side display device (CRT, etc.)
Adopts the display of the three primary colors of RGB, and (a) the colors that can be reproduced and displayed at once are reproduced by displaying the input color image by processing the intermediate gradation colors by a certain method by the dither method or the like. Alternatively, (b) reproduced display is performed based on the limited color obtained by selecting by an appropriate method using an LUT (look-up table).

As an improved invention using the dither method shown in (a) above, there is Japanese Patent Publication No. 64-5307. According to the present invention, an image frame is divided into two fields (so-called interlaced scanning), white display of a plurality of gradations is performed in the first field to prevent image flicker, and at the same time improve image brightness and resolution. This is a method of displaying colors during scanning of two fields, in which the luminance of each color separation signal of green, blue, and red is discriminated to activate the luminance.

As an improved invention using the LUT (look-up table) shown in (b) above, there is a dynamic limited color selection method filed on December 25, 1991 by the same applicant. This is because pixels of the input color image are arranged in the RGB space, the RGB space is appropriately divided according to the distribution density of the pixels arranged in the RGB space, and a pixel with a high frequency of use is selected. The LUT is created by selecting the limited colors required for reproduction.

According to this method, the most frequently used pixels are selected for each input color image by changing the size for dividing the RGB space according to the distribution density of the pixels of the input color image. Reproduction display can be performed based on an appropriately selected color. However, since the size to be divided for each input color image is appropriately determined, there are some problems in real-time display and high-speed processing.

Similarly, as a color image display system using an LUT, there is Japanese Patent Publication No. 3-52071. This is to calculate the popularity (number of pixels) included in the small space for the pixels arranged in the color space, and to calculate the relationship between the maximum registered popularity color and the simultaneous display color that has not yet been determined. This is a method of selecting simultaneous display colors depending on whether or not a prescribed fixed method is satisfied, and the color quantization with little image quality deterioration is performed by making the calculation of the popularity as coarse as possible and making the remaining pixels as uniform as possible. It is intended. However, since the input color image is divided by a predetermined small space size, colors that are unnecessary for reproduction are included at the stage of calculating the popularity,
There is a certain limit to the reproduction of an appropriate display color corresponding to the input image. Moreover, it is not suitable for real-time display because of the structural reason that it can be processed only after the target color image is input, and the complicated processing makes high-speed processing difficult.

[0007]

However, the conventional techniques of the dither method and the method using the LUT shown in the above (a) and (b) are aimed at the data of the color image input from the input device. Then, certain processing is performed to activate the colors required for reproduction, and the limited colors required for reproduction are selected. Therefore, it is necessary to perform a certain fixed process even for colors that are unnecessary for reproduction display included in the input color image, which slows down the processing speed and complicates the processing method. On the contrary, since the reproduced display color may not correspond to the input image, the reproduced display has a certain limit. The present invention has been made in consideration of the above circumstances, and the colors necessary for reproduction are extracted in advance before the method of selecting the limited color using the dither method or the LUT. It is intended to reproduce by a method such as a dither method or an LUT based on the color.

[0008]

In order to achieve the above object, the present invention provides a color image composed of a group of pixels,
The pixels are arranged on the basis of the RGB data in an RGB space which is a three-dimensional space in which the respective RGB colors are orthogonal coordinate axes, and the maximum gradation levels of the RGB components are Rm and G, respectively.
When m and Bm and the minimum gradation levels are R0, G0, and B0, respectively, the coordinates (R0, G0, B) in the RGB space are set.
All pixels of the color image including the group of pixels forming a substantially elliptical shape scattered around the achromatic line OF obtained from the position O point of (0) and the position F point of coordinates (Rm, Gm, Bm) Among these, the color image display method is characterized in that only the pixels forming the elliptical shape are extracted as the colors necessary for reproduction.

[0009]

In the color image including the group of pixels, the pixels are arranged in the RGB space based on the RGB data, and among all the pixels including the group of pixels forming the substantially elliptical shape scattered around the achromatic line OF, By extracting the pixels forming the elliptical shape, only the image necessary for reproduction is extracted in advance from the color image input from the input device, and a method using an LUT based on the extracted image data, Since a limited color is selected and reproduced by the dither method or the like, the image quality corresponding to the input color image can be efficiently reproduced.

[0010]

The present invention will be described in detail with reference to the drawings. As shown in FIG. 1, a color image composed of a group of pixels input from an input device (not shown) in a RGB space formed of a three-dimensional space having coordinate axes orthogonal to each of the RGB colors is used to generate the pixel 11 based on the RGB data. To place. When the maximum gradation level of each of the RGB components is m, the maximum value of all the components is represented by the position of the point F. Similarly, when all the components are at the minimum level, the position is at the point O, and the straight OF line represents an achromatic color.

The maximum gradation levels of the RGB components are respectively R
m, Gm, and Bm, and the minimum gradation levels are R0, G0, and B0, respectively.
, The coordinates in the RGB space (Rm, G
OF obtained from the coordinates (R0, G0, B0)
Line 13 is an achromatic line. This OF line 13
The pixels 11 forming the elliptical shape 15 are extracted from all the pixels including the group of the pixels 11 forming the substantially elliptical shape 15 scattered around the. Then, the input color image is reproduced and displayed by performing processing such as dither processing or selection of limited colors using the LUT based on the extracted pixels 11.

Next, a method of extracting the pixels forming the elliptical shape 15 will be described. Generally, when a color image input from a scanner or the like which is an input device is reproduced on a display device such as a CRT, a color particularly required is represented by a straight OF line 13 in an RGB space composed of RGB orthogonal coordinate axis systems. It is the color around the coloring line (see FIG. 1). That is, O
A set of pixels 11 at a position where an elliptical shape 15 formed of a pseudo ellipsoid centered on the F line 13 is formed is a set of colors necessary for reproduction on the display device.

Conversely, the pixels 11 that do not form the substantially elliptical shape 15 are colors that are not so important when reproduced on a display device such as a CRT. The above experience side is the premise of the present invention.

Therefore, a specific method for determining whether or not the pixel 11 is a pixel forming the substantially elliptical shape 15 will be described below.
First, when the color separation accuracy capability of an input color image is M colors, the maximum number of colors of each RGB represented when the maximum value level at which each RGB component can be color separated is ^m is 2 ^m.
= M _R , 2 ^m = M _G , 2 ^m = M _B , and the total number M of RGB combinations with the maximum resolution accuracy is M = M _R × M _G
It is expressed as × M _B = 2 ^{3 m} . The total number N of limited color combinations can be represented by N = 2 ⁿ .

The relation of the following equation is derived from M = 2 ^{3 m} and N = 2 ⁿ . Let n ÷ 3 = n ₁ remainder n _2, and consider p such that (1) n ₁ ≦ p <m. (2) Using the value of p thus obtained, each RGB component is appropriately divided to extract a group of pixels forming a substantially elliptical shape.

That is, each RGB component may be divided into 2 ^(mp) intervals and the limited color N may be selected from the number of combinations of RGB components P = 2 ^{3 (mp)} .

To actually select this limited color N, only one certain component has a value, such as (0, 0, d), out of the total number P of RGB component combinations divided by the value of p. , A combination in which the other two components are 0 is excluded. For example, the case of d = 2 ^(mp) is applicable. A combination such as (d ₁ , d ₂ , d ₃ ) d ₁ >> d ₂ , d ₃ in which only one component has a value and which is much larger than the other two components (the difference is d _x ).

In this way, the color of the combination to be removed is the pixel arranged outside the ellipsoidal shape 15, and the remaining color corresponds to the pixel arranged inside the elliptical shape 15.

By changing the value of p or changing the above condition of d _x , the limited color N can be automatically and rapidly selected by the automatic processing by the computer.

The above calculation means will be described using specific numerical values. First, the limited color N is 256 colors (N = 2 ⁿ =
2 ⁸ ), the total number of RGB combinations with maximum resolution accuracy is 2 for M colors
In the case of 60,000 colors (M = 2 ^{3 m} = 2 ^{3 · 6} ), m = 6, and n = 8 is obtained from the equation (1), so that n ₁ = 2 and n ₂ = 2 can be obtained.

Therefore, the range of p is 2 ≦ p from the above equation (2).
It is between <6 and can take values of p = 2, 3, 4, 5. If the value of p = 3 is selected, 2 ^(mp) = 2 ³ = 8
Number of RGB combinations at intervals P = 2 ^{3 (mp)} = 2 ^{3 · 3} = 512
You can get the individual.

That is, in the case of the number of combinations P of RGB, the limited color N (256 colors) is extracted by dividing into 64 ways (0, 8, 16, 24, ..., 504). ..

Of the RGB components thus obtained, for example, (0, 0, 8), (0, 0, 24)
8), (8, 8, 248) and other components are excluded, and (24
Components such as 8, 8, 8), (16, 24, 248) are excluded because only one component has a much larger value than the other two components (see above).

In this way, the substantially elliptical shape arranged in the RGB space is changed by appropriately changing the value of p within a predetermined range to change the division interval or by appropriately changing the difference (d _x ) between the RGB components. The pixels to be formed can be extracted.

In particular, in the embodiment, the color used in the multivalued dither method is selected in advance by the above method. At this time, the colors are not selected for each of the RGB components, but are selected according to the combination of RGB. Therefore, it is not possible to approximate the unselected colors (pixels outside the elliptical shape 15) for each of the RGB components. Can not. Therefore, in the present embodiment, a method of replacing each combination with a color (combination) having the shortest distance in the RGB orthogonal coordinate system is adopted.

[0026]

EFFECT OF THE INVENTION A color image made up of a group of pixels
The pixels are arranged in the B space based on the RGB data, and the pixels arranged inside and outside the elliptical shape are discriminated from the group of pixels having the substantially elliptical shape scattered around the achromatic line OF. , By extracting only the pixels located inside the elliptical shape, the image quality reproduced by the conventional method using the LUT or the multi-value dither method is improved, and the automatic processing by the computer is simple and the high-speed processing is possible. Since it is possible, almost real-time display is possible.

[Brief description of drawings]

FIG. 1 shows a group of pixels in which a color image input by an input device according to an embodiment of the present invention is arranged in an RGB space.

[Explanation of symbols]

 11 pixels 13 OF line 15 elliptical shape

Claims (1)

[Claims] 1. In a color image composed of a group of pixels, the pixels are arranged based on RGB data in an RGB space consisting of a three-dimensional space having coordinate axes orthogonal to each of the RGB colors, and the gradation maximum level of the RGB components is set. Each R
m, Gm, and Bm, and the minimum gradation levels are R0, G0, and B0, respectively.
In the above case, the coordinates in the RGB space (R0, G
0, B0) position O and coordinates (Rm, Gm, Bm) position F
Of all the pixels of the color image including the group of pixels scattered around the achromatic line OF obtained from points and forming a substantially elliptical shape, only the color required to reproduce the pixel forming the elliptical shape. A method for displaying a color image, characterized by extracting as.