JPH06289830A - Color selecting method of color image display system - Google Patents

Abstract

PURPOSE:To reproduce colors which are close to an original image by varying the number of divisions of a color space corresponding to an input image and then varying the size of a partial color space. CONSTITUTION:Which of (m) partial color spaces each pixel of input image data 27 is present in is calculated and the address (8 bits) of an entry of an LUT 22 wherein the representative color of the partial color space is set is written in the location of a frame memory 21 corresponding to the pixel. In the memory 21, display image data obtained by compressing 24-bit full color data to 8 bits are completed. Further, 8-bit data in the memory 21 are sent to the LUT 22 and the 24-bit representative color corresponding to the display pixel is outputted. Consequently, when the color distribution of the input image is concentrated on a narrow range, a maximum number of representative colors are obtained within the number of entries of the LUT 22, the number of divisions is automatically set to a large value to make the partial color space small, and consequently delicate differences in the colors of the input image can be reproduced.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color image display system for displaying an image using a look-up table, and more particularly, to more faithfully reproduce the color of an original image,
It relates to a method of setting color data in a look-up table.

[0002]

2. Description of the Related Art In a look-up table type color image display system, the color reproducibility of a display image with respect to an original image depends on what color is set in the look-up table. It will be.

Various methods have been proposed for selecting colors to be set in the look-up table. The simplest method is to sample a color space such as RGB space at equal intervals.

Here, the RGB color space means R (red),
It is a three-dimensional space of an orthogonal coordinate system in which coordinate axes representing G (green) and B (blue) components are orthogonal to each other.

In the above-described sampling at equal intervals, the RGB color space is divided into fixed sizes, and one color in each of the divided partial color spaces is divided into the partial color space. Is selected as a representative color of all the colors in the table and set in the look-up table. That is, the representative color is uniformly taken from the entire color space without considering the color distribution of the input image, and the look
It is set in the up table.

As a method of sampling in accordance with each input image, there is a method disclosed in Japanese Patent Laid-Open No. 62-8193, for example. This is because the lower bits of the RGB information of the original image are cut off, a histogram is taken with the values, and a value indicating a frequency higher than a predetermined ratio is adopted as a representative color, and a partial area of the look-up table is used. Set to. Then, by using the value obtained by cutting off the lower bits described above, colors having the same ratio of R, G, and B, that is, colors having the same chromaticity but different lightness are grouped, and a group having a large number of data is grouped. We have adopted many representative colors from
It is set in the remaining area of the table.

[0007]

In the method of sampling the entire color space at equal intervals, a representative color that is not necessary for some input images may be set in the look-up table. There is a problem that the color is reduced and the image quality is deteriorated.

Further, the method of selecting the representative color according to the input image, which is described in the above-mentioned publication, always divides the RGB color space into a fixed size regardless of the color distribution of the input image. Then, the color of one point in the divided partial color space is selected as the representative color of the pixel group existing in the partial color space. Therefore, even if the color distribution of the input image is biased, the size of the partial color space does not change, so that a large number of pixels are concentrated in a specific partial color space and the same representative color is obtained in those pixels. As a result, the color reproducibility of the display image with respect to the input image deteriorates.

[0009]

SUMMARY OF THE INVENTION The present invention is a system for converting an original image and writing it into a frame memory, and accessing a look-up table according to its contents to display a color image. When the number of the partial color spaces in which the pixels of the original image are present is m (m: integer), the m can be set in the look-up table. The number N of the partial color spaces is determined so that the number of the partial color spaces is equal to or less than the number of such partial color spaces, a representative color is selected from each of the m partial color spaces at that time, and the representative color is selected in the look-up table. The above problem is solved by setting.

Further, in a system in which an original image is converted and written in a frame memory and a look-up table is accessed according to the contents to display a color image, R, G and B axes of RGB color space are predetermined. Number n
A first step of dividing the RGB color space into n × n × n partial color spaces by dividing into (n: an integer) pieces; counting the number of the partial color spaces in which the pixels of the original image exist; A second step in which it is m (m: integer),
A third step of determining whether the number m is greater than the number of colors that can be set in the look-up table;
If it is determined to be large, a predetermined number is subtracted from n and the first to third steps are repeated. If it is determined to be small or equal, a representative color is selected from each of the m partial color spaces and the look The above problem is solved by having the fourth step of setting in the up table.

As the representative color, the color of the center point in each partial color space may be adopted.

[0012]

In the present invention, the number of divided color spaces changes according to the input image, and as a result, the size of the partial color space changes.

[0013]

FIG. 2 is a block diagram showing the structure of the present invention. In the figure, 24 is a CRT for displaying a color image, 2
A frame memory 1 holds display image data to be displayed on the CRT 24, and holds 8-bit pixel data for each display pixel to be displayed on the CRT 24. 22 is a look-up with 256 entries
It is a table (hereinafter referred to as LUT), and each entry has a data length of 8 bits for R, G, and B, for a total of 24 bits. Then, the color data set in the entry of the LUT 22 indicated by each pixel data of the frame memory 21 is converted into an analog signal by the DA converter 23 and displayed as a color image on the CRT 24.

Reference numeral 25 is an original image, and 26 is a scanner for scanning the original image 25 and outputting full color data in which each of R (red), G (green) and B (blue) components is composed of 8 bits. An image processing unit 20 receives the above-described full-color data as input image data 27, determines a representative color, sets the representative color in the LUT 22, and calculates data for accessing the LUT 22 for each pixel to calculate a frame. Write to memory 21. Next, the operation of this embodiment will be described with reference to the flowchart shown in FIG. 1 and FIGS. 3 to 5.

FIG. 4 shows the RGB color space. Each axis has a value from 0 to 255, and the input image data 27 in which each of R, G, and B has 8 bits is present in this space.

As shown in FIG. 1, the image processing unit 20 first sets the origin of the RGB color space to (0,0,0) and the farthest point to (255,255,255), and the input image data 27
To the RGB color space. Next, the initial value of the division number n is set to 256, and the R, G, and B axes are equally divided into n to divide the entire RGB color space into n × n × n (= N) partial color spaces. To do.

Then, of the divided partial color spaces, the number of partial color spaces in which the pixels of the input image are present is counted and the value is set to m. If this m is larger than 256, which is the number of entries in the LUT 22, the value of n is decremented by 1 and the RGB color space is equally divided again by this new n.

Thus, while the value of n is decreased by 1, the division of the RGB color space and the counting of the number m of partial color spaces in which the pixels of the input image exist are repeated. That is, the size of the partial color space is gradually increased from the state of representing the full-color original image as it is, while the number of partial color spaces that include the input image data therein is reduced. It represents going.

When m becomes equal to 256 or becomes smaller in time, a representative color is selected from the m partial color spaces in which the pixels of the input image are present, and the representative colors are sequentially input to the LUT 22, as shown in FIG. Set. This representative color adopts the color at the center point of each partial color space, and each of R, G, and B has a data length of 8 bits, that is, a total of 24 bits.

When the selection of the representative color is completed in this way, the pixel data to be written in the frame memory 21 is calculated next.

That is, for each pixel of the input image data 27, of the m partial color spaces described above,
The address (8 bits) of the entry in the LUT 22 in which the representative color of the partial color space is set is calculated in which space it exists, and written in the position corresponding to the pixel in the frame memory 21. .

When the above processing is applied to all the pixels,
In the frame memory 21, display image data obtained by compressing 24-bit full color data into 8 bits is completed.

Then, as schematically shown in FIG. 3, the 8-bit data on the frame memory 21 is converted into the LUT 22.
The 24-bit representative color corresponding to the display pixel is output from the entry of the LUT 22 pointed to by the 8-bit data, converted into an analog signal by the DA converter 23, and the color image is displayed on the CRT 24. Is displayed.

As described above, in the present invention, the size of the partial color space to be divided is variable. Even if the total number of the partial color spaces becomes larger than the number of entries in the LUT 22 due to the fine division, only the partial color space in which the pixels of the input image exist is actually related to the selection of the representative color. is there. Therefore, when the color distribution of the input image is concentrated in a narrow range, the division number n is automatically set to a large value so that as many representative colors as possible can be obtained within the number of entries of the LUT 22. The color space is reduced, and as a result, subtle color differences in the input image are reproduced.

In this embodiment, the color at the center point of the partial color space is used as the representative color. However, for example, a point other than the center point is closest to the origin (0,0,0) of the partial color space. Colors can also be adopted.

[0026]

According to the present invention, when the colors (pixels) of the original image are dispersed, the number of divisions is reduced to increase the partial color space, and when concentrated, the number of divisions is increased. The partial color space can be made smaller, and the representative color is selected only from the partial color space in which the pixel exists.
Colors closer to the original image can be reproduced.

[Brief description of drawings]

FIG. 1 is a flowchart illustrating a processing procedure of an image processing unit according to an embodiment.

FIG. 2 is a block diagram showing the configuration of the color image display system of the embodiment.

FIG. 3 is an explanatory diagram illustrating a method of displaying a color image using a look-up table.

FIG. 4 is a diagram showing an RGB color space and pixel distribution.

FIG. 5 is a schematic diagram showing how a partial color space and a representative color are selected.

[Explanation of symbols]

 20 image processing unit 21 frame memory 22 look-up table 24 CRT 27 input image data

Claims (3)

[Claims] 1. In a system in which an original image is converted and written in a frame memory, and a look-up table is accessed according to the contents to display a color image, N (N: integer) partial colors are used as a color space. Divide into spaces equally,
Let m be the number of partial color spaces in which the pixels of the original image exist.
(M is an integer), the number m of the partial color spaces is determined so that m is equal to or less than the number of colors that can be set in the look-up table, and the number N of the partial color spaces is determined. A color selecting method in a color image display system, wherein a representative color is selected from each of m partial color spaces and set in the look-up table. 2. A system in which an original image is converted and written in a frame memory, and a look-up table is accessed according to the contents of the original image to display a color image.
The RGB color space is divided into n × n by dividing each R, G, B axis of the GB color space into a predetermined number n (n: integer).
A first step of dividing into × n partial color spaces, a second step of counting the number of the partial color spaces in which the pixels of the original image exist and setting it as m (m: integer), the number m Is larger than the number of colors that can be set in the look-up table, and if it is larger, the predetermined number is subtracted from n and the first to third steps are repeated. And a fourth step of selecting a representative color from the m partial color spaces and setting the same in the look-up table when it is determined that they are smaller or equal to each other. Color selection method. 3. The color selection method in a color image display system according to claim 1, wherein the representative color is a color of a center point in each partial color space.