JPH0571099B2 - - Google Patents

Description

[Detailed Description of the Invention] [Summary] When converting the gradation of a multi-gradation color image such as a full-color natural image, the specific color area specified by the image creator must be converted to the color of that area or the artificial color. For a color image area other than the specific color area, which is displayed in a composite color, a predetermined number of specified colors representing the area are automatically determined, and the gradation of the area is automatically converted using these specified colors. Thereby, it is possible to efficiently perform gradation conversion of a color image while suitably reflecting the intention of the image creator.

[Industrial application field]

The present invention relates to a gradation conversion control method for converting a multi-gradation color image such as a full-color natural image into a color image having an arbitrarily specified number of color gradations smaller than the number of color gradations.

[Conventional technology]

Multi-gradation color images such as full-color natural images output from video cameras, video tapes, etc. have a huge number of color gradations, for example, ²²⁴ colors.

On the other hand, in color image processing systems, the number of color gradations is small (for example, 16 to 256 colors) due to limitations due to the characteristics of terminal devices, the amount of information is compressed, the amount of hardware is reduced, and the processing time is shortened. In many cases, there are restrictions on

Conventionally, as a tone conversion control method for converting a color image with an extremely large number of gradations into a color image with a small number of gradations, a black and white image is input in advance from a black and white television, camera, etc. A black-and-white image is divided into image parts corresponding to specific areas or specific brightness level ranges, and the image creator selects the display color for each image part from among the specified colors, and artificially creates a color image close to the input color image. This method inputs a multi-gradation color image from a color TV, camera, etc., and performs automatic gradation conversion to obtain a color image close to the original input color image within a predetermined number of colors. The method of doing so, etc. was used.

[Problem that the invention seeks to solve]

In conventional gradation conversion control methods, it is difficult to artificially express natural colors with a small number of specified colors, and the image creator inputs each area and the specified color for each area. For,
The problem was that it was time consuming. On the other hand, in the method (2), since the color is automatically determined, even if it is desired to arrange a desired specific color for a specific area, this is not possible.

[Means for solving problems]

The means taken by the present invention to solve the above-mentioned problems in the conventional gradation conversion control system will be explained with reference to FIG.

FIG. 1 is a block diagram showing the structure of the present invention.

In FIG. 1, reference numeral 110 denotes a specific color area extracting means that extracts at least one specific color area displayed in an arbitrary specific color from the color image area. The specific color area is manually designated by the image creator. The specific color is a color within a specific color area or an artificially synthesized color, and is arbitrarily specified by the image creator.

Reference numeral 120 denotes designated color determining means, which automatically determines a designated color at a predetermined location representative of a color image area other than the specific color area.

Reference numeral 130 denotes a specific color input means for inputting each specific color for each specific color area.

Reference numeral 140 denotes a gradation converting means, which automatically determines, for areas other than the specific color area, areas to be displayed with each specified color determined by the specified color determining means 120, and for each specific color area, Each specific color input from the specific color input means 130 is input.

[Effect]

The operation of the present invention having the configuration shown in FIG. 1 will be explained with reference to FIG. 2. FIG. 2 is an explanatory diagram of an input color image of the present invention.

In the input color image 200 of FIG.
0 is the sky, 221-223 are clouds, 230 is the sea, 24
0 is a sandy beach, 251 is a leaf, 252 is a tree trunk, 2
60 is a person, 270 is an island, 280 is a yacht, 29
is a ship.

Now, assume that the multi-tone color image 200 is toned-converted into a color image with fewer k colors.

The image creator selects r (k>
A specific area to be displayed in a specific color where r≧1) is manually specified. Extraction of a specific color area is performed by specifying a specific color from among the colors of the color image 200 as a point or area and extracting an area of these colors at the same time, or by specifying a specific area. .

For example, in Fig. 2, the number of designated colors k = 16, the number of specific colors r = 4, and among these four specific colors, 2 sky
10. For the sea 230 and the sandy beach 240, the colors in each area are used, and the color of the leaves 251 is artificial green. In this case, the sky 210,
When each pixel in each area of the sea 230 and the sandy beach 240 is the same, each area 210, 23
By specifying any one point between 0 and 240, the sky 210 and the sea 230 are created as areas with that color.
and each area of the sandy beach 240 is specified. If, for example, the color of the sky 210 is not a single color and you cannot specify the color of that area by simply specifying one point, specify an arbitrary closed area 211 within the sky 210 and select the closed area. The color of the sky 210 is specified by the color 211.

The closed area of the tree leaf 251 for which artificial greenery is specified is specified.

The designated color determining means 120 automatically determines a predetermined number of designated colors representing the area, ie, k-r (=12) colors, for areas other than the specific color area.
This method of automatically determining the specified color can be performed using various known methods.

The specific color input means 130 inputs each specific color for each specific color area.

For areas other than the specific color area, the gradation converting means 140 automatically determines areas to be displayed with each specified color determined by the specified color determining means 120. For each specific color area, each specific color input from the specific color input means is input.

As described above, the specific area specified by the image creator can be displayed in an artificial specific color or a specific color in a color image as desired by the image creator, and other areas can be displayed in each color. The gradation can be automatically converted and displayed using a color that represents the area.

〔Example〕

An embodiment of the present invention will be described with reference to FIGS. 2 to 5.

FIG. 3 is a block diagram of the configuration of an embodiment of the present invention, FIG. 4 is an illustration of a color lookup table used in the embodiment, and FIG. 5 is an explanation of the clustering method used in the embodiment. It is a diagram.

(Description of Configuration) In FIG. 3, specific color area extraction means 110,
The specified color determining means 120, the specific color input stage 130, and the gradation converting means 140 are as described in FIG.

150 is a television camera that outputs full-color natural image information. An image input control unit 160 performs A/D conversion on full-color natural image information and writes it into the image memory 170. The image memory 170 generally stores multi-tone color image information, such as A/D converted full-color natural images.

In the specific color region extraction means 110, 111 is an input color lookup table (input color LUT) creation section. As shown in Figure 4, the input color LUT converts each pixel of a color image into a color image.
Represented by entry number, each color entry has red (R), green (G), and blue (B) that indicate the color information of the corresponding pixel.
Each data is stored. Reference numeral 112 is a color appearance frequency table creation unit in which appearance frequencies for each color are stored. Both of these input color LUTs and color appearance frequency tables are used in color image processing, and their creation methods are well known.

Reference numeral 113 denotes an area input means into which coordinate information of a specific color area to be displayed in a specific color is input. As the area input means 113, for example, a known tablet board is used.

Reference numeral 114 denotes a specific color specifying means, which specifies the color portion of the specific color area when the specific color area is displayed with the color of the area. For example, a tablet board is used as the specific color specifying means, and the specific color is specified by indicating a desired specific color portion in the specific color area with a light pen.

Reference numeral 115 denotes a region extracting unit which extracts from the color appearance frequency table 112 the color information of the specific color region input from the region input stage 113 or the color information of the specific color specified by the specific color specifying means 114. Extract color regions. The extracted specific color information is added to the specific color input means 130, and the specific color area information is added to the specified color determination means 120.

In the specific color input means 130, 131 is a composite specific color input stage, which inputs an artificially synthesized specific color for displaying a specific color area. A specific color selection section 132 selects a color for displaying a specific color area. That is, when the display color of a specific color area is a color in the specific color area, the specific color information inputted from the area extraction unit 115 in response to the input operation from the area input means 113 or the specific color designation means 114 is used. If the artificial composite color is the display color, select the composite color input from the composite specific color input means 131,
It is added to the gradation conversion means 140. These operations are
This is done manually by the image creator.

The gradation conversion means 140 has an output color lookup table (output color LUT) 14 inside.
1, for a specific color area, gradation conversion is performed using a specified artificially synthesized specific color or a specific color (r color) in the specific color area, and for areas other than the specific color area, the specified color is used. Gradation conversion is performed using the (k-r) colors determined by the determining means 120.

180 is a display unit that displays the input color image and the tone-converted color image.

(Explanation of operation) The operation shown in Fig. 3 is performed using the input error image shown in Fig. 2.
An example will be described in which gradation conversion is performed using 16 designated colors, 4 of which are specific colors, that is, gradation conversion is performed using k=16 colors and r=4 colors. Here, the sky 210,
The four areas of the sea 230, the sandy beach 240, and the leaves 251 are set as specific color areas. Among these, since the sky 210 is not a single color, the color of the closed area 211 in the sky 210 is used as the specific color, and the sea 230 and the sandy beach 240 are respectively Since it is a single color, the color of point P in Umi 230,
In the sandy beach 240, the color of point Q is designated as a specific color for each area, and the leaves 251 are displayed in an artificially synthesized "green" color. Other areas are
It shall be displayed in 12 designated colors that are automatically determined to represent those areas.

The full color natural image information input from the television camera 150 is input to the image input control section 160.
After being D-converted, it is stored in the image memory 170 as multi-tone color image information.

The input color LUT creation unit 111 creates a fourth color LUT based on the multi-tone color image information in the image memory 170.
Create an input color LUT as shown in the figure.

The color appearance frequency table creation unit 112 creates a color appearance frequency table from the input color LUT.

The image creator inputs the closed regions of the sky 211 and the leaves 215 using the region input means 113, and specifies the P point of the sea 230 and the Q point of the sandy beach 240 using the specific color specifying means 114, and selects a specific color for each region. Specify.

The area extraction unit 115 extracts the color information of the specific color area input from the area input unit 113 and the color information of each specific color specified by the specific color designation unit 114 from the color appearance frequency table. By extracting each color information in each specific color area, each specific color area, that is, each area of the sky 210, the sea 230, the sandy beach 240, and the leaves 251 is extracted. Each extracted specific color information is added to specific color input means 130, and specific color area information is added to specified color determination means 120.

To the designated color determining means 120, each color and its appearance frequency information are added from the color appearance frequency table creation section 112, and each specific color region information is added from the region extraction section 115. The specified color determining means 120
Based on the color information and appearance frequency information obtained by removing the color information of each specific color from each color and its appearance frequency information,
The remaining k-r (=12) designated colors representing these colors are automatically determined. As a result, 12 designated colors representing areas other than the specific color area are automatically determined.

This automatic determination of the designated color can be performed, for example, by a known clustering method.

FIG. 5 explains the principle of the clustering method. The color information of each pixel of a color image is expressed as Y
When expressed as a (luminance) signal, a Q signal, and an I signal, each pixel is divided into multiple groups (clusters) in a three-dimensional coordinate space consisting of the Y, Q, and I axes, as shown in Figure 5. It is distributed in the form of If you specify the number of clusters and perform clustering, the specified number of clusters will be automatically formed. The color information of the centroid of each cluster is the color information representing each cluster. Therefore, in the case of this example, the number of clusters is specified as 12 and clustering is performed.
By obtaining the color information of each center of gravity of the 12 clusters formed, a specified color representing an area other than the specific color area is automatically determined. It was decided in this way
The 12 specified colors are applied to the gradation conversion means 140.

The image creator inputs green color for displaying the tree leaves 251 using the composite specific color input means 131. Furthermore, when the closed region information of the leaf 251 is input from the region input means 1113, the image creator uses the specific color selection section 132 to select that color instead of the color information of the leaf 251 input from the region extraction section 115. Composite specific color input means 131 at the entry number
Select the green color input from gradation conversion means 1.
Add to 40. Sky 210 from area input means 113
When the closed region 211 of
The color information of the closed area 211 input from 15 is selected and added to the gradation conversion means 140. Even when the specific color specifying means 114 specifies the colors of points P and Q as specific colors for the sea 230 and the sandy beach 240, the color information of points P and Q input from the area extraction unit 115 is selected. , is added to the gradation conversion means 140.

The gradation converting means 140 converts the sky 2 which is a specific color area.
10, the sea 230, the sandy beach 240, and the leaves 251 are r input from the specific color selection section 132.
(=4) Each specific color is the color of the corresponding specific color area, and for areas other than the specific color area, k determined by the specific color determining means 120 is used.
- Perform gradation conversion using r (=12) colors.

Specifically, the color output LUT 141 in the gradation conversion means 140 is divided into four specific colors and the remaining 12 colors.
Rewrite the color with the specified color. That is, in the case of an area other than the specific color area, color information for the specified color of the cluster is stored in all color entries corresponding to the color entry numbers included in the cluster of each of the 12 specified colors. In addition, in the case of a specific color area, the color entry for each specific color area's color entry number is
Stores color information for each corresponding specific color.

The tone-converted color image information output from the tone conversion means 140 is added to the color display section 180 and displayed in color. Note that when gradation conversion is not performed, the contents of the color input LUT created by the color input-LUT creation section 111 are input as they are to the color output LUT 141, so the color display section 180 displays the input color image. is also displayed.

If you want to modify or change the gradation-converted color image, you can modify or change the specified number of colors k, the specific color area, the content of the specific color, etc. as appropriate and perform the above-mentioned processing. good.

Although one embodiment of the present invention has been described above,
The invention is not limited to this example.

For example, to extract a specific area or determine a specified color,
In addition to the methods shown in the above embodiments, various known methods can be used.

〔Effect of the invention〕

As explained above, according to the present invention, the following effects can be obtained.

(b) The image creator can arbitrarily select a specific color area and display that specific color area using the colors in the input color image as they are or using any artificial composite color. , it is possible to perform gradation conversion into a color image that satisfactorily reflects the intentions of the image creator.

(b) For areas other than specific color areas, a predetermined number of specified colors that best represent them can be automatically determined, so that the overall color gradation conversion can be performed while reflecting the image creator's intentions well. can be done efficiently and quickly.

[Brief explanation of the drawing]

Figure 1...Block explanatory diagram of the configuration of the present invention, Figure 2
Figure...Explanatory diagram of the input color image of the present invention, Figure 3...
FIG. 4 is an explanatory diagram of an embodiment of the present invention; FIG. 4 is an explanatory diagram of a color lookup table used in the embodiment; FIG. 5 is an explanatory diagram of a clustering method used in the embodiment. In FIG. 1, 110...specific color area extraction means, 120... specified color determining means, 130... specific color inputting means, 140... gradation conversion means.

Claims (1)

[Scope of Claims] 1. In a gradation conversion control method for converting a multi-gradation color image into a color image having an arbitrarily specified number of color gradations smaller than the number of color gradations, (a) a color image; a specific color region extracting means 110 for extracting at least one specific color region displayed in any specific color from within the region; (b) regarding color image regions other than the specific color region;
(c) Specific color input means 130 that inputs a predetermined specific color for the specific color area; (d) Specific color area. For other areas, the area to be displayed with each specified color determined by the specified color determination means 120 is automatically determined, and for the specified color area, the specified color input from the specified color input means 130 is automatically determined. A gradation conversion control method characterized by comprising: gradation conversion means 140 for receiving input.