JPH10164377A - Color number reduction processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accelerate a processing and to perform the color number reduction processing of high quality while reducing a required storage capacity by individually performing the color number reduction processing to object image data expressed in the three-dimensional space of RGB by respective dimension units and returning them to three dimensions after the processing. SOLUTION: As one example, in the case of color-reducing the full-color images of about 16,700,000 colors to about 32,000 colors, based on the inputted full-color images D1, a graph (histogram) for indicating a using frequency to respective gradations is prepared for the respective components of RGB (S1) and conversion to 32-gradation gray scales D31-D33 is performed based on the result of the histogram (S2.) The prepared 32-gradation gray scale images D31-D33 of the respective components of RGB are synthesized into one image again (S3.) In this case, the image of the maximum of about 32,000 colors of 32×32×32=32,768 (colors) is attained and the color number reduction processing is ended.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color reduction processing apparatus, and more particularly to a color reduction processing apparatus for reducing the color of an original image.

[0002]

2. Description of the Related Art A color reduction processing apparatus for reducing the color of a full-color image expressed in a three-dimensional space of RGB without deteriorating the quality thereof, displaying image data on a display device, and increasing the speed when printing with a printing device is known. It's being used. Conventionally, as this type of color reduction processing apparatus, one disclosed in Japanese Patent Application Laid-Open No. 8-87256 is known.

In the color reduction processing apparatus disclosed in the publication, R
The RGB space of a full-color image composed of (red), G (green), and B (blue) is divided into predetermined gradation units, and color reduction processing is performed according to the number of gradations used in each subdivided block. I have.

FIG. 4 is a diagram showing this color reduction method. The image to be reduced in the drawing has 64 gradations for each of the RGB colors, and has a total expression number of 64 × 64 × 64 = 262,144 (about 260,000 colors). Here, assuming that RGB is a three-dimensional space and the gradation unit to be divided is 8 gradations, the RGB space is divided into 8 × 8 × 8 = 512 (block). Therefore, the number of colors that can be represented by each subdivision block is 8 × 8 × 8 = 512 (colors). Therefore, how many colors are used in each of the 512 divided blocks, and how many pixels are used in each color in the color reduction target image are counted. Select as many as you want. Thereby, the color reduction processing of the full-color image is realized.

[0005]

The above-described conventional color reduction processing apparatus has the following problems. First, when an electronic circuit is to be realized, a high-speed and expensive arithmetic unit (CPU) and a large-capacity and expensive storage memory are required. This is because the color reduction processing is performed in a three-dimensional RGB space, and it is necessary to process the color reduction target image in three-dimensional RGB. As a result, many working areas are required in a storage memory or the like, and much time is required for processing itself.

On the other hand, when the software is realized by an arithmetic device such as a computer, much time is required for processing, and a large working area is required in a storage memory or the like. This is because the image data must be held as a three-dimensional array of RGB at the time of processing, and the color reduction processing itself needs to be processed by three-dimensional calculation. Therefore, in order to perform the color reduction processing, a large amount of work area is required in a large-capacity storage memory, and the processing itself requires much time.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and provides a color reduction processing apparatus capable of performing high-quality color reduction processing while increasing the processing speed and reducing the required storage capacity. With the goal.

[0008]

In order to achieve the above object, the invention according to claim 1 performs color reduction processing on image data to be subjected to color reduction represented in a three-dimensional space of RGB individually for each dimension, and after the processing. It is configured to return to three dimensions again.

In the invention according to claim 1 configured as described above, the color space of a full-color image represented in three dimensions is converted to one dimension to perform color reduction processing. That is, the color reduction processing can be performed at higher speed by individually performing the color reduction processing for each dimension. For example, when performing color reduction processing, RG
Since all colors included in the B color space must be referred to or operated, if the number of gradations of each of the RGB components of the original image is 64, 64 × 64 × 64 = 262,144 (Times) will be calculated as one unit. However, by performing this operation in one dimension, the operation is performed with 64 + 64 + 64 = 192 (times) as one unit, so that a significant increase in speed can be realized.

When the color reduction processing is implemented as software, the storage capacity required for conversion can be reduced. For example, in the case of a conventional one, RGB of image data
Since each component needs to be stored, the storage capacity required when performing the color reduction processing on the original image is three times the size of the original image. However, if the color reduction processing is performed individually for each dimension as described above, since the RGB components are converted separately, the storage capacity required for the color reduction processing is only the size of the original image. The storage capacity is one-third that of the conventional technology.

According to a second aspect of the present invention, in the color reduction processing apparatus according to the first aspect, the color reduction processing includes:
The overall frequency of use of the gradation colors used in the color reduction target image data is detected, and processing is performed with priority given to this frequency of use. In performing the color reduction processing, the total use frequency of the gradation colors used in the color reduction target image data is detected, and the processing is performed with priority given to this use frequency. That is, the number of colors used in the color reduction target image and the frequency of use of each color are checked to create a histogram graph, and the processing according to the histogram lowers the quality of the image after the color reduction processing. Is preventing.

Further, according to a third aspect of the present invention, in the color reduction processing apparatus according to the second aspect, a gradation color which is not adopted due to priority of use frequency is converted into an approximate gradation color. If tone colors are selected in descending order of frequency of use by priority processing according to the histogram, other tone colors will not be selected, but this is approximated to the closest tone color. As a result, it is possible to secure a frequently used gradation in the color reduction target image.

Further, a fourth aspect of the present invention is the color reduction processing apparatus according to any one of the first to third aspects, wherein the apparatus is configured to be repeated a predetermined number of times. By repeating the color reduction process several times, for example, a 256-color image is generated from a full-color image. The process of narrowing down little by little in the order of frequency of use is repeated.

[0014]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a color reduction processing apparatus according to an embodiment of the present invention. In the figure, when performing color reduction processing, a full-color image 101 input to the apparatus is input to a color reduction processing unit 102, where the color reduction processing is performed. Specific processing of the color reduction processing unit 102 will be described later. The color reduction processing unit 102 generates a color-reduced image 103 as an output, and generates the color-reduced image 1
03 is input to the RGB image control unit 104. The RGB image control unit 104 outputs the reduced color image to the display device 105 and the printing device 106.
Is displayed on the display based on the color-reduced image 103, and the printing device 106 performs printing processing on paper based on the color-reduced image 103, and ends.

On the other hand, when it is not necessary to reduce the color of the input image, the full-color image 101 input to the device is
The image is directly output to the display device 105 and the printing device 106 by the B image control unit 104, and the process ends.

Next, the operation of this embodiment having the above configuration will be described. As an example, a case in which a full-color image of about 16.7 million colors is reduced to about 32,000 colors will be described.
FIG. 2 is a flowchart corresponding to the color reduction processing of the color reduction processing unit 102 shown in FIG. Based on the full-color image D1 input in step S1, the color reduction processing unit 102 creates a graph (histogram) representing the frequency of use for each gradation for each of the RGB components.

Next, the input full-color image D1 is
256 gradation gray scale images D21 to D21 of RGB components
D23, and based on the result of the histogram described above in step S2, a 32-tone gray scale D3
1 to D33.

The process of reducing the color from the result of the histogram in step S2 to a 32-tone gray scale will be described with reference to FIG. In step S2, first, 2
The 56-tone histogram is equally divided into eight regions for each color. This state corresponds to H1. Next, the area of each of the divided regions is calculated. This calculation is performed based on the following equation. The area of each region = the width of each region × the sum of the frequency of use of each gradation in each region. When the area of each region is obtained in this manner, 32 gradations are assigned to each of the eight regions based on the area ratio.

In accordance with the allocated number, the gradation is selected in the order of the frequency of use in each area, and the other gradations are approximated to the gradation closest thereto. As a result, it is possible to secure a frequently used gradation in the color reduction target image. This state corresponds to H2. That is, a histogram is generated for each component of RGB three-dimensional data, the histogram is divided into eight small areas, priorities are determined, and 32 gradations are assigned based on the priorities. Therefore, the color reduction processing is performed individually for each component.

Further, the assignment of 32 gradations is performed with priority given to the overall frequency of use based on the histogram, so that the deterioration of the image is minimized. Returning to FIG. 2, the 32-gradation grayscale images D31 to D33 of the respective RGB components created as described above are output once again in step S3.
Combine into two images. In this case, the image becomes 32 × 32 × 32 = 32,768 (color), which is a maximum of about 32,000 colors, and the color reduction processing ends.

In the embodiment described above, about 1,67
Although the case where a full-color image of 10,000 colors is reduced to about 32,000 colors has been described, there is no limitation on the number of colors of the original image and the number of colors after the reduction. Further, a color image of about 32,000 colors is generated by reducing the 256-tone grayscale image to 32 tones, but the number of tones after conversion is not limited. For example, it is possible to reduce the color image of about 32,000 colors to a color image of 6 × 6 × 6 = 216 (colors) by using a gray scale of 6 gradations instead of 32 gradations.

On the other hand, the color reduction process shown in FIGS. 2 and 3 may be repeated several times. That is, first, a 256-tone grayscale image is converted into a 32-tone grayscale image, and then the same processing is repeated to convert the 32-tone grayscale image into an 8-tone grayscale image. As a result, a color image of 256 colors can be generated. More specifically, in the flowchart, the gradation conversion may be designated by a parameter, and the color reduction processing unit 102 may be called in a nested manner.

In this way, although the processing time increases, high-quality color reduction can be performed more easily. As described above, a histogram is created for each component in the RGB three-dimensional space (step S1), and a 32 gradation gray scale is generated based on the area ratio of each region while dividing the histogram into predetermined gradation units (step S1). Since S2), the color reduction processing can be performed in one dimension, the color can be reasonably reduced by combining the colors at the end, the color reduction processing can be realized at a high speed with a smaller storage capacity, and the color reduction processing is used for the color reduction target image data. Since the processing according to the number of colors and the frequency of use of each color is performed, it can be said that this is a high-quality color reduction processing.

[0024]

As described above, the present invention provides a color reduction processing apparatus capable of performing high-speed color reduction processing by converting a three-dimensional color space into one dimension and performing color reduction processing. can do. In addition, storage resources and the like can be reduced. That is, it can be realized with high speed, high quality, and a smaller storage capacity. According to the second aspect of the present invention, since the color reduction processing is performed based on the total use frequency of the gradation colors, it is possible to prevent the image quality from deteriorating.

Further, according to the invention of claim 3,
By converting an unselected gradation color into an approximate color, a gradation color frequently used in the color reduction target image can be secured. Further, according to the invention of claim 4, since the gradation colors to be used are gradually narrowed down, the processing time is increased, but high-quality color reduction can be performed without difficulty.

[Brief description of the drawings]

FIG. 1 is a block diagram of a color reduction processing apparatus according to an embodiment of the present invention.

FIG. 2 is a flowchart executed by a color reduction processing unit.

FIG. 3 is a diagram schematically illustrating the concept of gradation conversion in a color reduction process.

FIG. 4 is a diagram illustrating the concept of a conventional color reduction process.

[Explanation of symbols]

 Reference Signs List 101 Full-color image 102 Color reduction processing unit 103 Color-reduced image 104 RGB image control unit 105 Display device 106 Printing device

Claims (4)

[Claims] 1. A color reduction processing apparatus characterized in that color reduction target image data expressed in a three-dimensional space of RGB is individually subjected to color reduction processing in units of dimensions, and is returned to three-dimensional after processing. 2. The color reduction processing apparatus according to claim 1, wherein the color reduction processing detects a total use frequency of a gradation color used in the color reduction target image data, and determines the use frequency. A color reduction processing device characterized by performing processing with priority. 3. The color reduction processing device according to claim 2, wherein a gradation color not adopted due to priority of use frequency is converted into an approximate gradation color. 4. The color reduction processing apparatus according to claim 1, wherein the processing is repeated a predetermined number of times.