JPH09261502A - Color converter - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a color converter by which color expression close to an original image is attained by assigning lots of representative color to an area with a high frequency of occurrence of image data and with a high frequency of colors and assigning representative colors whose number corresponds to the frequency of occurrence even to an area with a low frequency of occurrence. SOLUTION: A color space conversion means 32 converts picture element data configuring a color image into a coordinate system of a color space. A color space division means 33 divides the color space into plural space blocks, a color frequency count means 34 counts frequency of color of picture element, data for each space block. A used color number assignment means 35 assigns the used color number to each space block at a rate in response to the frequency of color. A representative color decision means 36 decides the representative colors by the used color number assigned as above. A color conversion means 37 converts the picture element data in each space block into each representative color.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color image processing apparatus, and more particularly to a color image processing apparatus for converting a full color image into a color image of an arbitrary finite color.

[0002]

2. Description of the Related Art When a full-color image represented by three primary color signals is output to a finite color image output device having a limited number of colors and color types, such as a multi-color printer or multi-color display, a full-color image is output. It is necessary to convert a full-color image into a finite-color image in advance by quantizing, i.e., color-converting the used color into a fixed representative color that can be reproduced by the image output device. In conventional color conversion, a full-color image is first converted into a histogram in a three-dimensional space of RGB, and this three-dimensional space is divided into a sufficient number of spatial blocks. Then, determine the color with the highest occurrence frequency in each spatial block as the representative color of the spatial block,
The number of colors can be reduced to the number of spatial blocks by converting all the pixels in the spatial block into the representative color, that is, by performing color conversion.

However, in the above-described conventional color conversion method, the representative color is evenly assigned to each spatial block even if there is a large difference in the color frequency and the occurrence frequency of pixels between the spatial blocks. For this reason, in the high-frequency area, there is a problem in that, for example, red and vermilion are converted into the same red without being distinguished. Further, the pixel near the boundary of each spatial block is converted into the representative color of the spatial block to which the pixel belongs, even if it is close to the color of the adjacent spatial block, which makes it difficult to faithfully represent the color.

In order to solve such a problem, for example, in Japanese Unexamined Patent Publication No. 4-287570, a target block is identified based on a histogram of the target block and 26 surrounding blocks (for three dimensions) surrounding the target block. It is determined whether or not the block is a concentrated block (block in which pixels are frequently generated).
When the block of interest is determined to be a concentrated block, it is then determined whether the block of interest is a concentrated block of the same color as the surrounding blocks. Assign a representative color,
If it is determined that the blocks are concentrated blocks of the same color, the same representative color as that of the surrounding blocks is assigned to the block of interest to reduce the number of colors, that is, perform color conversion.

[0005]

However, in the above-mentioned conventional technique, only one representative color is assigned to the space in which the concentrated blocks of the same color are dense, so that the color changes subtly. There is a problem that the fidelity with respect to the original image is lowered because each color is expressed in the same color even in the existing area. Furthermore, since the representative color is not assigned to the block in which the frequency of occurrence of pixels is low, there is a problem that the color infrequently occurs is omitted and the fidelity to the original image is also low.

An object of the present invention is to solve the above-mentioned problems of the prior art, and to allocate a large number of representative colors to regions where pixel occurrence frequency or color frequency is high, while assigning many representative colors to regions where the frequency is low. An object of the present invention is to provide a color conversion device capable of performing color expression close to an original image by allocating a corresponding number of representative colors.

[0007]

Means for Solving the Problems In order to achieve the above object, the present invention is characterized in that the following means are taken. (1) A color space conversion unit that converts each pixel data forming a color image into a coordinate system of a color space, a color space division unit that divides the color space into a plurality of space blocks, and pixel data of each space block. Color frequency counting means for counting the color frequency, used color number assigning means for assigning the used color number to each spatial block at a ratio according to the color frequency, and representative for the assigned used color number for each spatial block A representative color determining means for determining a color and a color converting means for converting the pixel data in each spatial block into each of the representative colors are provided. (2) Color space conversion means for converting each pixel data forming a color image into a coordinate system of a color space, color space division means for dividing the color space into a plurality of space blocks, and pixel data for each space block. Occurrence frequency counting means for counting the occurrence frequency, use color number assigning means for assigning the use color number to each space block at a ratio according to the occurrence frequency, and representative for the use color number assigned for each space block A representative color determining means for determining a color and a color converting means for converting the pixel data in each spatial block into each of the representative colors are provided.

According to each of the above-mentioned configurations, a large number of representative colors are assigned to an area having a high color frequency (or occurrence frequency) of image data, and a color frequency (occurrence frequency) is assigned to an area having a low color frequency (occurrence frequency). ) Is assigned a representative color in a ratio according to.
For this reason, delicate color expression can be performed in a region with high color frequency (occurrence frequency), while color loss can be prevented even in a region with low color frequency (occurrence frequency) and color representation faithful to the original image can be achieved. .

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail with reference to the drawings. FIG. 1 is a functional block diagram of a multicolor printer which is an embodiment of the present invention. The image reading means 31
The color original 1 is read in pixel units and each pixel is converted into image data of electric signals. The color space conversion means 32 converts each pixel data forming a color image into a color space coordinate system. The color space dividing means 33 divides the color space into a plurality of space blocks. The color frequency counting means 34 counts the color frequency of pixel data for each spatial block. The used color number assigning unit 35 assigns the used color number to each spatial block at a ratio according to the color frequency. The representative color determination means 36 determines the representative colors for the allocated number of used colors for each spatial block. The color conversion means 37 converts the pixel data in each spatial block into each of the representative colors.

Next, the operation of the embodiment of the present invention will be described in detail with reference to the flowchart of FIG. In step S101, the color image 1 is read by the image reading unit 31.
Are read in pixel units. In step S102, color space conversion is performed by the color space conversion means 32, and each pixel data forming the color image is converted into the coordinate system of the color space. In step S103, color space division is performed by the color space division means 33 to divide the color space into a large number of spatial blocks. In the present embodiment, it is assumed that the color space is first divided into 16 spatial blocks.

Steps S104 to S106 are processing executed by the color frequency counting means 34. In step S104, a histogram relating to the number of colors is created in the color space, and in step S105, each space is calculated based on the histogram. The number of colors in a block, that is, the number of pixels having different colors in each block is counted. In step S106, a color frequency table showing the color frequency of each spatial block is created based on the counting result and the total number of colors.

FIG. 3 is a diagram showing an example of the color frequency table,
It is shown that 30% of the total number of colors appears in the space block 1 and 15% of the total number of colors appears in the color space block 2. It also indicates that no color appears in the color space block 16, that is, no pixel exists.

In step S107, the used color number assigning means 35 provisionally determines the number of colors assigned to each spatial block based on the color frequency table of FIG. Less than,
A method of determining the number of assigned colors will be described with reference to FIG.

FIG. 4 is a diagram showing a calculation method and calculation result of the number of assigned colors for each spatial block. The number of assigned colors is the number of printable colors (here, 16) in the color frequency (%).
It is calculated by multiplying (color). For example, since the color frequency of the spatial block 1 is 30%, the value "4" obtained by multiplying this by the number of printable colors (16) (fractions after the decimal point are truncated) is assigned to the spatial block 1. Becomes a number. Similarly, two colors are provisionally assigned to the space block 2 and one color is provisionally assigned to the space block 3. As in the case of the space blocks 6 to 15, if the number of assigned colors is “0” when the number after the decimal point is rounded down, one color is assigned,
When the color frequency is 0% as in the spatial block 16, the assigned color number is also "0".

By the way, in the above-described allocation method, the total number of assigned colors does not always match the number of printable colors. Also in this embodiment, the total number of colors that can be assigned is 19 while the number of colors that can be printed is 16.
Therefore, in subsequent step S108 and subsequent steps, a process of reducing the number of assigned colors by the difference between the total number of assigned colors and the number of printable colors is performed.

Steps S108 to S111 are the processes executed by the representative color determining means 36. In step S108, the maximum frequency of each spatial block is calculated based on the color space histogram created in step S104. Each color is provisionally determined as a representative color of each spatial block. At this time, in the spatial blocks to which a plurality of colors are assigned like the spatial blocks 1 and 2,
The representative colors for the assigned number of colors are determined in order from the upper color of the histogram.

In this way, the originally assigned number of colors (1
When more than six colors (19 colors) of representative colors are provisionally determined, in step S109, all the distances between the respective representative colors on the color space are obtained, and the color space as shown in FIG. A correlation diagram is created. In the color space correlation diagram, in the column where reference numeral 7 and reference numeral 8 intersect, the representative color represented by reference numeral 7 (for example,
The distance in the color space between the representative color of the color space block 3) and the representative color represented by reference numeral 8 (for example, the representative color of the color space block 4) is recorded.

When the distances between all the representative colors are obtained, in step S110, only the differences (three pairs in this embodiment) are selected from among the combinations of the representative colors having the shortest distance among them.
By integrating the representative color of the spatial block on the low color frequency side into the representative color of the spatial block on the high color frequency side for each set, each of the representative colors is reduced by a total of three colors. Note that the method of integrating the two representative colors into one representative color is not limited to the above, and each representative color may be integrated into an intermediate color between the two, or each representative color may have a color frequency of both. Alternatively, the intermediate colors may be integrated in a ratio corresponding to.

As described above, the 16 representative colors are determined, and when the number of colors to be assigned to each space block is finally determined, the space to which a plurality of colors are assigned is determined in step S111. The block is further divided by the number of assigned colors, and the representative color of each spatial block is determined. In step S112, the color conversion unit 37 converts the pixels included in each spatial block into the representative color of the spatial block.

As described above, in the present embodiment, the distribution of the representative colors in the color space is changed according to the color frequency of the pixel, and a large number of representative colors are assigned to the area having high color frequency of the image data. Since a representative color is assigned to a region having a low color frequency at a rate according to the color frequency, a large number of representative colors are assigned to a region having a high color frequency, which enables delicate color expression. Color loss is prevented, and color expression faithful to the original picture is possible.

In the above embodiments, the number of representative colors assigned to each spatial block is determined according to the color frequency of image data. However, the present invention is not limited to this and the number of image data It may be determined according to the frequency of occurrence.

[0022]

According to the present invention, the following effects are achieved when outputting a full-color image to a multicolor printer or a multicolor display in which the number of colors and color types are limited. (1) According to the invention of claim 1, a large number of representative colors are assigned to the area having high color frequency of the image data, and the representative color is assigned to the area having low color frequency at a ratio according to the color frequency. Therefore, a delicate color expression can be performed in an area having a high color frequency, while a color loss can be prevented even in an area having a low color frequency, and a color expression faithful to the original image can be performed. (2) According to the invention of claim 2, a large number of representative colors are assigned to the area where the image data occurrence frequency is high, and the representative color is also assigned to the area where the occurrence frequency is low at a rate according to the occurrence frequency. Therefore, while the subtle color expression can be performed in the area where the occurrence frequency is high, the color loss can be prevented even in the area where the occurrence frequency is low, and the color expression faithful to the original image can be performed.

[Brief description of drawings]

FIG. 1 is a functional block diagram of a printer that is an embodiment of the present invention.

FIG. 2 is a flowchart showing the operation of FIG.

FIG. 3 is a diagram showing an example of a color frequency table.

FIG. 4 is a diagram showing a calculation method and a calculation result of an assigned color number.

FIG. 5 is a diagram showing an example of a color space correlation diagram.

[Explanation of symbols]

31 ... Image reading means, 32 ... Color space converting means, 33 ... Color space dividing means, 34 ... Color frequency counting means, 35 ... Used color number assigning means, 36 ... Representative color determining means, 37 ... Color converting means

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁶ Identification code Agency reference number FI Technical display location H04N 1/46 H04N 1/46 Z

Claims (2)

[Claims] 1. A color space conversion means for converting each pixel data forming a color image into a coordinate system of a color space, a color space division means for dividing the color space into a plurality of space blocks, and a pixel for each space block. Color frequency counting means for counting the color frequency of the data, used color number assigning means for assigning the used color number to each spatial block at a ratio according to the color frequency, and the used number of assigned colors for each spatial block And a color conversion unit that converts pixel data in each spatial block into each of the representative colors. 2. A color space conversion means for converting each pixel data forming a color image into a coordinate system of a color space, a color space division means for dividing the color space into a plurality of space blocks, and a pixel for each space block. Occurrence frequency counting means for counting the occurrence frequency of data, use color number assigning means for assigning the use color number to each spatial block at a ratio according to the occurrence frequency, and the use color number for each space block And a color conversion unit that converts pixel data in each spatial block into each of the representative colors.