JPH11252385A - Color character discriminating method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a color character discriminating method capable of accurately discriminating the color character. SOLUTION: This color character discriminating method comprises the discriminating the color of a character image from the values of three input data of RGB with respect to a pixel which is discriminated to be a character image by an image attribute discriminating processing of a character, a dot, a photograph, etc. In this case, the method comprises a difference minimum value discriminating step S6 for obtaining a difference minimum value as a minimum value within the difference of arbitrary two data of RGB data, an RGB minimum value discriminating step 7 of obtaining an RGB minimum value as a minimum value within RGB data values and color discriminating steps S8 to S18 of discriminating the color of a character image, based on the RGB minimum value and the difference minimum value.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color character discrimination method for discriminating a color of a character image.

[0002]

2. Description of the Related Art When an original such as a map, a drawing, a document, a print or a photographic paper is read by a scanner or the like and output to a printer or a PC (personal computer), an adaptive process especially for the sharpness of characters is performed. Is being done. In this adaptive processing, not only the distinction between black characters and colored characters, but also the color characters are further distinguished into cyan, magenta, yellow, red, green, and blue colors, thereby further improving character sharpness. Becomes possible.

[0003] A conventional color character discriminating method will be described below.
The relationship between the GB determination bit and each color will be shown (Table 1) and will be described with reference to FIG. FIG. 17 is a flowchart showing a conventional color character determination method.

[0004]

[Table 1]

In FIG. 17, first, multi-gradation image data (character image data) whose characters have been determined is input (S11).
0), the three RGB data of the target pixel of the image data are read (S111). Next, the R data is compared with a threshold value (the central value 128 of the input 8-bit data) (S11).
2) If the R data is equal to or greater than the threshold value, "1" is inserted into the R determination bit (S113), and if the R data is less than the threshold value, "0" is inserted into the R determination bit (S11).
4). Next, the G data is compared with the threshold value (S11).
5) If the G data is equal to or larger than the threshold, "1" is set to the G determination bit (S116), and if the G data is smaller than the threshold, "0" is set to the G determination bit (S11).
7). Further, the B data is compared with the threshold value (S11).
8) If the B data is equal to or larger than the threshold, "1" is set to the B determination bit (S119), and if the B data is smaller than the threshold, "0" is set to the B determination bit (S12).
0).

Here, referring to (Table 1), the colors of the characters cyan, magenta, yellow, red, green, and blue are determined (S121). For example, when the determination bit of R is “1”, the determination bit of G is “0”, and the determination bit of B is “0”, it is determined that the character is a red character.

As described above, the color character discrimination processing of the target pixel is completed (S122), the RGB data of the next target pixel is read (S111), and the above processing (steps 111 to 122) is repeated. When the color character discrimination processing for all pixels is completed (S122), the color character discrimination processing ends.

[0008]

However, in the above-described conventional color character discrimination method, the threshold value is a fixed value of the median value, and the influence of the character density and the threshold value when discriminating the character color is determined. There is a problem in that the accuracy of character discrimination is large and low.

In this color character discriminating method, it is required that color discrimination be performed with high accuracy in color character discrimination.

It is an object of the present invention to provide a color character discriminating method capable of discriminating a character color with high accuracy.

[0011]

In order to solve this problem, a color character discriminating method according to claim 1 of the present invention comprises a character,
A color character discrimination method for discriminating a color of a character image from values of three input data of RGB for a pixel determined to be a character image by image attribute determination processing such as a halftone dot or a photograph.
A difference minimum value determining step of obtaining a difference minimum value as a minimum value of a difference between any two data of the B data;
An RGB minimum value determining step of determining an RGB minimum value as a minimum value among the GB data values; and a color determining step of determining a color of a character image based on the RGB minimum value and the difference minimum value. I have.

As a result, a color character discriminating method capable of discriminating the character color with high accuracy is obtained.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The invention according to claim 1 of the present invention is a method of determining three input data of RGB for a pixel determined to be a character image by image attribute determination processing of characters, halftone dots, photographs, and the like. A color character discriminating method for discriminating a color of a character image from a value, a difference minimum value discriminating step of obtaining a difference minimum value as a minimum value of a difference between arbitrary two data of RGB data, And an RGB minimum value determining step of determining an RGB minimum value as a minimum value among the above, and a color determining step of determining a color of a character image based on the RGB minimum value and the difference minimum value. The density and the threshold value are not used for color discrimination, and there is an effect that the character density and the threshold value have no effect.

According to a second aspect of the present invention, a color of a character image is determined for a pixel determined as a character image by image attribute determination processing of a character, a halftone dot, a photograph, or the like from the values of three input data of RGB. A color character discriminating method for discriminating, wherein a difference minimum value discriminating step of obtaining a difference minimum value as a minimum value of a difference between arbitrary two data of RGB data; and RGB as a maximum value of RGB data values. RGB to find the maximum value
A maximum value determining step; and a color determining step of determining a color of the character image based on the RGB maximum value and the minimum difference value, wherein the character density and the threshold value are not used for the color determination. Has the effect that there is no influence by the character density or the threshold value.

According to a third aspect of the present invention, a color of a character image is determined for a pixel determined to be a character image by image attribute determination processing of a character, a halftone dot, a photograph, or the like based on three input data values of RGB. An RGB maximum value determining step of obtaining an RGB maximum value as a maximum value of RGB data, and an RGB minimum value determining step of obtaining an RGB minimum value as a minimum value of RGB data. And R
An RGB median discriminating step of finding an RGB median as a median of the GB data; comparing the RGB median with a threshold value; and a RGB maximum value, an RGB minimum value, and an RGB value.
A color discrimination step of discriminating the color of the character image based on the median value and the threshold value. Is also reduced.

According to a fourth aspect of the present invention, in the third aspect of the present invention, the threshold value is set to an RGB maximum value and an RGB value.
A step of determining a threshold value obtained from the minimum value is provided between the RGB median value determining step and the color determining step, and the threshold value depends on the RGB maximum value and the R value.
This is only the GB minimum value, and has the effect that the dependency of the threshold value is reduced as compared with the conventional case.

According to a fifth aspect of the present invention, the color of a character image is determined for pixels determined to be a character image by image attribute determination processing for characters, halftone dots, photographs, etc., based on the values of three RGB input data. An RGB maximum value determining step of obtaining an RGB maximum value as a maximum value of RGB data, and an RGB minimum value determining step of obtaining an RGB minimum value as a minimum value of RGB data. And R
An RGB median discriminating step of obtaining an RGB median as a median of the GB data; a difference calculating step of obtaining a difference between the RGB median and the RGB minimum value; comparing the RGB median with a threshold value; And a threshold value, and a color determining step of determining the color of the character image based on the RGB maximum value, the RGB minimum value, the RGB median value, the difference, and the threshold value. The value dependence is only the RGB median value and the difference, and has the effect that the dependence on the threshold value is reduced as compared with the conventional case.
This has the effect that color character discrimination independent of the value character density is performed.

According to a sixth aspect of the present invention, a color of a character image is determined for a pixel determined to be a character image by image attribute determination processing of a character, a halftone dot, a photograph, or the like from three RGB input data values. An RGB maximum value determining step of obtaining an RGB maximum value as a maximum value of RGB data, and an RGB minimum value determining step of obtaining an RGB minimum value as a minimum value of RGB data. And R
An RGB median determining step for obtaining an RGB median as a median of the GB data; a difference calculating step for obtaining a difference between the RGB median and an RGB maximum value; comparing the RGB median with a threshold value; And a threshold value, and a color determining step of determining the color of the character image based on the RGB maximum value, the RGB minimum value, the RGB median value, the difference, and the threshold value. The value dependence is only the RGB median value and the difference, and has the effect that the dependence on the threshold value is reduced as compared with the conventional case.

Hereinafter, an embodiment of the present invention will be described with reference to FIG.
This will be described with reference to FIG. (Embodiment 1) A color character discrimination method according to Embodiment 1 of the present invention will be described with reference to FIGS. FIG. 1 is an explanatory diagram of a color judgment of a yellow character in the color character judgment method according to the first embodiment of the present invention, and FIG. 2 is an explanation of a color judgment of a blue character in the color character judgment method according to the first embodiment of the present invention. FIG. 3 and FIG. 3 are explanatory diagrams of the color discrimination of the cyan character in the color character discriminating method according to the first embodiment of the present invention, and FIG. 4 is a description of the color discrimination of the red character in the color character discriminating method according to the first embodiment of the present invention. FIG. 5 and FIG. 5 are explanatory diagrams of the color determination of the magenta character in the color character determination method according to the first embodiment of the present invention, and FIG. 6 is a description of the color determination of the green character in the color character determination method according to the first embodiment of the present invention. FIG. FIG. 7 is a flowchart showing a color character discrimination method according to the first embodiment of the present invention.

In FIG. 7, first, image data (character image data) for which a character is determined is input (S1), and R data 1, G data 2, and B data 3 of a target pixel of the image data are read (S2). . Here, the difference 4 between the R data 1 and the G data 2 is obtained (S3). Next, a difference 5 between the G data 2 and the B data 3 is obtained (S4). Further, a difference 6 between the B data 3 and the R data 1 is obtained (S5). Here, the minimum difference 7 among the differences 4, 5, and 6 determined above is determined (S 6, minimum difference determination step), and the RGB minimum 8 among the R data 1, G data 2, and B data 3 is calculated. (Or maximum value) (S7, RGB minimum value determination step (RG
B maximum value determination step)). Difference minimum value 7 and difference 4
(S8), and when the minimum difference value 7 is equal to the difference 4, the RGB minimum value 8 (or the maximum value) is compared with the B data 3 (S9).

1 and 2, if it is assumed that the value of the B data 3 exists between the values of the R data 1 and the G data 2, the difference 4 does not become the minimum difference value 7. Therefore, when the difference 4 is equal to the minimum difference value 7, the value of the B data 3 does not exist between the values of the R data 1 and the G data 2, and the difference 4 is the difference 4 among the differences 4, 5, and 6. Are the minimum values of the R data 1 and the G data 2
It is obviously larger or smaller than. The ideal values of RGB for 8-bit yellow are:
The R data is 255, the G data is 255, and the B data is 0. In addition, ideal blue RGB values are 0 for R data, 0 for G data, and 255 for B data. Therefore, the difference 4 between the R data 1 and the G data 2 is the minimum difference value 7
, The character becomes a yellow character or a blue character depending on the value of the B data 3.

In FIG. 7, when the RGB minimum value 8 (or the maximum value) is equal to the value of the B data 3, it is determined that the character is a yellow character (or a blue character) (S10).
When the GB minimum value 8 (or the maximum value) is not equal to the value of the B data 3, it is determined that the character is a blue character (yellow character) (S11). Next, when the difference minimum value 7 is not equal to the difference 4 (S8), the difference minimum value 7 is compared with the difference 5 (S12), and when the difference minimum value 7 is equal to the difference 5, RG
The B minimum value 8 (or the maximum value) is compared with the R data 1 (S13).

3 and 4, if it is assumed that the value of the R data 1 exists between the values of the G data 2 and the B data 3, the difference 5 does not become the minimum difference value 7. Therefore, when the difference 5 is equal to the minimum difference value 7, the value of the R data 1 does not exist between the values of the G data 2 and the B data 3, and the difference 5 is the same as the difference 4, the difference 5, and the difference 6. Is smaller than the G data 2 and the B data 3, so that the value of the R data 1 is clearly larger or smaller. The ideal cyan RGB values at 8 bits are R
The data is 0, the G data is 255, and the B data is 255. The ideal red RGB values are 255 for R data, 0 for G data, and 0 for B data. Therefore,
When the difference 5 between the G data 2 and the B data 3 is the minimum difference 7, the character becomes a cyan character or a red character depending on the value of the R data 1.

In FIG. 7, when the RGB minimum value 8 (or the maximum value) is equal to the value of the R data 1, it is determined that the character is a cyan character (or a red character) (S14).
If the B minimum value 8 (or the maximum value) is not equal to the value of the R data 1, it is determined that the character is a red character (cyan character) (S15). Next, when the minimum difference 7 is not equal to the difference 5 (S12), the minimum difference 7 is equal to the difference 6.

5 and 6, if it is assumed that the value of the G data 2 exists between the values of the R data 1 and the B data 3, the difference 6 does not become the minimum difference value 7. Therefore, when the difference 6 is equal to the minimum difference value 7, the value of the G data 2 does not exist between the values of the R data 1 and the B data 3, and the difference 6 is one of the differences 4, 5, and 6. Is smaller than the R data 1 and the B data 3, so that the value of the G data 2 is clearly larger or smaller. The ideal magenta RGB values at 8 bits are:
The R data is 255, the G data is 0, and the B data is 255. The ideal green RGB values are 0 for R data, 255 for G data, and 0 for B data. Therefore, the difference 6 between the R data 1 and the B data 3 becomes the minimum difference value 7
, The character becomes a magenta character or a green character depending on the value of the G data 2.

Here, in FIG. 7, the RGB minimum value 8 (or the maximum value) is compared with the value of the G data 2 (S16).
If the RGB minimum value 8 (or maximum value) is equal to the value of the G data 2, it is determined that the character is a magenta character (or green character) (S 17), and the RGB minimum value 8 (or maximum value) of the G data 2 If not equal to the value, it is determined that the character is a green character (magenta character) (S18).

Thus, the color character discrimination processing of the target pixel is completed (S19), the RGB data of the next target pixel is read (S2), and the above processing (S2 to S19) is repeated. If the color character discrimination processing for all the pixels for which character discrimination has been completed (S19), the color character discrimination processing ends. In addition,
Steps 8 to 18 constitute a color determination step.

As described above, according to the present embodiment, RG
Difference minimum value discrimination step S6 for finding the minimum difference value as the minimum value of the difference between any two data of the B data.
An RGB minimum value determining step (RGB maximum value determining step) S7 for obtaining an RGB minimum value (RGB maximum value) as a minimum value (or maximum value) of the RGB data values;
Color discriminating step S8 to discriminate the color of the character image based on the RGB minimum value (or the RGB maximum value) and the difference minimum value
Since S18 is provided, the character density and the threshold value are not used for color determination, and the influence of the character density and the threshold value can be eliminated, so that the character color can be determined with high accuracy. .

(Embodiment 2) A color character discrimination method according to Embodiment 2 of the present invention will be described with reference to FIGS. FIG. 8 is an explanatory diagram of a color character discrimination method according to the second embodiment of the present invention, and FIGS. 9 and 10 are flowcharts showing a color character discrimination method according to the second embodiment of the present invention.

In FIG. 9, first, image data for which a character has been identified is input (S21), and R data 9, G data 10, and B data 11 of a target pixel of the image data are read (S22). Here, as shown in FIG. 8, the RGB maximum value 12, the RGB minimum value 13, and the RGB median value 14 among the R data 9, the G data 10, and the B data 11 are obtained (S2).
3, S24, S25). The RGB maximum value 12 is compared with the value of the R data 9 (S26), and the RGB maximum value 12 is R
If it is equal to the value of the data 9, "1" is inserted into the determination bit 15 of R (S27). If the RGB maximum value 12 is not equal to the value of the R data 9 (S26), the RGB maximum value 1
2 and the value of the G data 10 are compared (S28). RGB
When the maximum value 12 is equal to the value of the G data 10, "1" is set to the G determination bit 16 (S29). If the RGB maximum value 12 is not equal to the value of the R data 9 or the value of the G data 10 (S26, S28), "1" is set to the B determination bit 17 (S30). Next, the RGB minimum value 13 and R
The value is compared with the value of data 9 (S31), and the RGB minimum value 13
Is equal to the value of the R data 9, "0" is inserted into the determination bit 15 of R (S32). If the RGB minimum value 13 is not equal to the value of the R data 9 (S31), the RGB minimum value 1
3 is compared with the value of the G data 10 (S33). RGB
When the minimum value 13 is equal to the G data 10, "0" is set to the G determination bit 16 (S34). If the RGB minimum value 13 is not equal to the value of the R data 9 or the value of the G data 10 (S31, S33), "0" is set to the B determination bit 17 (S35).

As shown in FIG. 10, the RGB median value 1
4 is compared with the value of the R data 9 (S36). If the RGB median value 14 is equal to the value of the R data 9, the R data 9 is compared with the threshold value 18 (the middle value 128 of the input 8-bit data). Then, when the value of the R data 9 is less than the threshold value, “0” is set to the determination bit 15 of R (S38).
When the value of the R data 9 is equal to or larger than the threshold value 18 (S3
7) Put “1” in the determination bit 15 of R (S3)
9). When the RGB median value 14 is not equal to the value of the R data 9 (S36), the RGB median value 14 is compared with the value of the G data 10 (S40). When the RGB median value 14 is equal to the value of the G data 10, the value of the G data 10 and the threshold value 1
8 (S41), and when the value of the G data 10 is less than the threshold value 18, "0" is set to the G determination bit 16 (S42). When the value of the G data 10 is equal to or larger than the threshold value 18 (S41), "1" is set to the G determination bit 16 (S43). When the RGB median value 14 is not equal to the value of the R data 9 or the value of the G data 10 (S36) (S4)
0), the value of B data 11 is compared with threshold value 18 (S
44) If the value of the B data 11 is less than the threshold value 18, "0" is set to the B determination bit 17 (S45). When the value of the B data 11 is equal to or greater than the threshold value 18 (S4
4), "1" is put into the determination bit 17 of B (S4).
6).

Ideal cyan RG with 8-bit data
The values of B are 0 for R data, 255 for G data, 255 for B data, magenta for R data for 255, G data for 0, B data for 255, and yellow for R data for 255, G data. The data is 255 and the B data is 0. In addition, ideal red RGB values are 255 for R data,
G data is 0, B data is 0, green is R data is 0, G data is 255, B data is 0, and blue is
R data is 0, G data is 0, and B data is 255. So this ideal cyan, magenta, yellow,
Table 1 shows binarized values of red, green, and blue RGB. Determination bit 15 of R obtained above
From the determination bit 16 of G and the determination bit 16 of G and the determination bit 17 of B, (Table 1) is referred, and cyan, magenta, yellow, red,
Green and blue characters are determined (S47). For example,
When the determination bit 15 of R is “1”, the determination bit 16 of G is “0”, and the determination bit 17 of B is “0”, the character is a red character. Thus, the color character discrimination process of the target pixel is completed (S48), the RGB data of the next target pixel is read (S22), and the above processes (S22 to S48) are repeated. If the color character discrimination processing of all pixels is completed (S4
8), the color character discrimination processing ends.

As described above, according to the present embodiment, RG
R for finding RGB maximum value as maximum value of B data
RGB maximum value determination step S23, RGB minimum value determination step S24 for obtaining an RGB minimum value as a minimum value of RGB data, and RGB median value determination step S25 for obtaining an RGB median value as a median value of RGB data
And a color discriminating step of comparing the RGB median value with a threshold value and discriminating the color of the character image based on the RGB maximum value, the RGB minimum value, the RGB median value, and the threshold value.
47, the threshold value depends only on the median value, and the threshold value dependency can be reduced as compared with the related art.

(Embodiment 3) A color character discrimination method according to Embodiment 3 of the present invention will be described with reference to FIGS. FIG. 11 is an explanatory diagram of a color character discrimination method according to the third embodiment of the present invention. FIGS. 12 and 13 are flowcharts showing a color character discrimination method according to the third embodiment of the present invention.

Steps 51 to 65 in FIG. 12 showing the color character discrimination method according to the present embodiment correspond to steps 21 to 35 in FIG. 9 and step 67 in FIG.
To 77 correspond to steps 36 to 46 in FIG.
Steps 78 and 79 in FIG. 12 correspond to steps 47 and 48 in FIG. 9, and therefore description thereof is omitted. That is, the color character discrimination method according to the present embodiment is different from the second embodiment in that it has step 66. In step 66, a threshold is obtained from the following equation (1) using a and b in FIG.

Threshold value = (RGB minimum value × a + RGB maximum value × b) ÷ (a + b) (1) As the threshold value, the value shown in equation (2) can be used.

Threshold value = RGB minimum value × α + RGB maximum value × β + γ (2) where 0 ≦ α + β ≦ 1, | γ | <20 (experimental value), and equation (2) Fine adjustment is performed on the threshold value of Expression (1) by γ.

As described above, according to the present embodiment, the threshold value determination step 66 for obtaining the threshold value from the RGB maximum value and the RGB minimum value is performed by the RGB median value determination step and the color determination step in the second embodiment. Between the RGB maximum value and RG
Only the B minimum value, and the dependency of the threshold value is reduced as compared with the conventional case.

(Embodiment 4) A color character discrimination method according to Embodiment 4 of the present invention will be described with reference to FIGS. FIG. 14 is an explanatory diagram of a color character discrimination method according to the fourth embodiment of the present invention. FIGS. 15 and 16 are flowcharts showing a color character discrimination method according to the fourth embodiment of the present invention. FIG. 1 shows a color character discrimination method according to the present embodiment.
Steps 81 to 95 in FIG. 5 correspond to steps 21 to 35 in FIG. Therefore, steps 96 to 109 will be described.

From steps 92, 94, and 95, the process proceeds to step 96. In step 96, as shown in FIG. 14, difference data 39 between the RGB median value 34 and the RGB minimum value 33 (or the RGB maximum value 32) is obtained. Next, the RGB median value 34 is compared with the value of the R data 29 (S97). When the RGB median value 34 is equal to the value of the R data 29, the difference data 39 and the threshold value 38 (=
30 (from an experimental value)) (S98), and when the obtained difference data 39 is less than the threshold value 38, "0" (or "1") is inserted into the determination bit 35 of R (S9).
9). When the obtained difference data 39 is equal to or larger than the threshold value 38 (S98), "1" (or "0") is set in the determination bit 35 of R (S100). RGB median 34
Is not equal to the value of the R data 29 (S97), RG
The B median 34 is compared with the value of the G data 30 (S10).
1). When the RGB median value 34 is equal to the value of the G data 30, the difference data 39 obtained above is compared with the threshold value 38 (S102).
If it is less than 8, "0" (or "1") is set in the G determination bit 36 (S103). When the difference data 39 is equal to or larger than the threshold value 38 (S102), "1" (or "0") is set in the G determination bit 36 (S102).
104). RGB median 34 is G for both R data 29
When the value is not equal to the value of the data 30 (S97) (S10
1) Compare the obtained difference data 39 with the threshold value 38 (S105), and when the obtained difference data 39 is less than the threshold value 38, set the determination bit 37 of B to “0” (or “1”). (S106). When the obtained difference data 39 is equal to or greater than the threshold value 38 (S105), "1" (or "0") is inserted into the B determination bit 37 (S105).
107).

Ideal cyan RG with 8-bit data
The values of B are 0 for R data, 255 for G data, 255 for B data, magenta for R data for 255, G data for 0, B data for 255, and yellow for R data for 255, G data. The data is 255 and the B data is 0. In addition, ideal red RGB values are 255 for R data,
G data is 0, B data is 0, green is R data is 0, G data is 255, B data is 0, and blue is
R data is 0, G data is 0, and B data is 255. So this ideal cyan, magenta, yellow,
Table 1 shows binarized values of red, green, and blue RGB. Determination bit 25 of R obtained above
From the determination bits 26 of G and G and the determination bits 27 of B (see Table 1), cyan, magenta, yellow, red,
Green and blue characters are determined (S108). For example, the determination bit 35 of R is “1”, the determination bit 36 of G is
Is "0" and the B determination bit 37 is "0", it is a red character.

Thus, the color character discrimination processing of the target pixel is completed (S109), the RGB data of the next target pixel is read (S82), and the above processing (S82) to (S109) are repeated. If the color character discrimination processing for all the pixels is completed (S109), the color character discrimination processing ends.

As described above, according to the present embodiment, RG
R for finding RGB maximum value as maximum value of B data
RGB maximum value determination step S83, RGB minimum value determination step S84 for obtaining an RGB minimum value as a minimum value of the RGB data, and RGB median value determination step S85 for obtaining an RGB median value as a median value of the RGB data
Calculating a difference between the RGB median value and the RGB minimum value (or the RGB maximum value) in step S96;
B color comparing step of comparing the median value with the threshold value, comparing the difference with the threshold value, and determining the color of the character image based on the RGB maximum value, the RGB minimum value, the RGB median value, the difference and the threshold value. Since steps S86 to S95 and S97 to S109 are provided, the dependency of the threshold value is only the RGB median value and the difference, and the dependency of the threshold value can be reduced as compared with the related art.

[0044]

As described above, according to the color character discriminating method according to the first aspect of the present invention, pixels which are discriminated as character images by image attribute discriminating processing of characters, halftone dots, photographs, etc. A color character discrimination method for discriminating the color of a character image from the values of three input data of RGB.
A difference minimum value determining step of obtaining a minimum difference value as a minimum value between the two data differences; an RGB minimum value determining step of obtaining an RGB minimum value as a minimum value of the RGB data values; A color discriminating step of discriminating the color of the character image based on the minimum value, the character density and the threshold are not used for the color discrimination, and the influence of the character density and the threshold on the color discrimination is eliminated. Therefore, the advantageous effect that the color of a character can be determined with high accuracy can be obtained.

According to the second aspect of the present invention, a pixel determined to be a character image by image attribute determination processing of a character, a halftone dot, a photograph, or the like is applied to a pixel of a character image from three RGB input data values. A color character determination method for determining a color,
A difference minimum value determining step of obtaining a difference minimum value as a minimum value of a difference between any two data of the data;
By including an RGB maximum value determining step of determining an RGB maximum value as a maximum value of the B data values, and a color determining step of determining a color of a character image based on the RGB maximum value and the difference minimum value, the character Since the density and the threshold value are not used for the color discrimination and the influence of the character density and the threshold value on the color discrimination can be eliminated, the advantageous effect that the color of the character can be discriminated with high accuracy can be obtained. .

According to the third aspect of the present invention, a pixel determined to be a character image by image attribute determination processing of a character, a halftone dot, a photograph, or the like is applied to a pixel of a character image based on three RGB input data values. A color character determination method for determining a color,
RG for calculating RGB maximum value as maximum value of data
A B maximum value determining step, an RGB minimum value determining step of obtaining an RGB minimum value as a minimum value of the RGB data, an RGB median determining step of obtaining an RGB median value as a median of the RGB data, By having a color discrimination step of comparing the median value with the threshold value and discriminating the color of the character image based on the RGB maximum value, the RGB minimum value, the RGB median value, and the threshold value, the dependence of the threshold value is reduced. Only the median value is obtained, and an advantageous effect that the dependency of the threshold value can be reduced as compared with the related art can be obtained.

According to the invention set forth in claim 4, according to claim 3,
In the invention described in (1), the step of determining a threshold value from the maximum RGB value and the minimum RGB value is a threshold value determining step.
By providing between the median value discriminating step and the color discriminating step, the dependence of the threshold value on the RGB maximum value and the RGB
Only the minimum value is obtained, and an advantageous effect that the dependency of the threshold value can be reduced as compared with the related art can be obtained.

According to the fifth aspect of the present invention, a pixel determined as a character image by image attribute determination processing of a character, a halftone dot, a photograph, or the like is applied to a pixel of a character image based on three RGB input data values. A color character determination method for determining a color,
RG for calculating RGB maximum value as maximum value of data
A B maximum value determining step, an RGB minimum value determining step of obtaining an RGB minimum value as a minimum value of the RGB data, an RGB median determining step of obtaining an RGB median value as a median of the RGB data, Median and R
A difference calculating step for obtaining a difference from the minimum GB value;
B color comparing step of comparing the median value with the threshold value, comparing the difference with the threshold value, and determining the color of the character image based on the RGB maximum value, the RGB minimum value, the RGB median value, the difference and the threshold value. Has the advantageous effect that the dependence of the threshold value is only the RGB median value and the difference, and the dependence of the threshold value can be reduced as compared with the related art.

According to the sixth aspect of the present invention, a pixel determined to be a character image by image attribute determination processing of a character, a halftone dot, a photograph, or the like is applied to a pixel of a character image from three RGB input data values. A color character determination method for determining a color,
RG for calculating RGB maximum value as maximum value of data
A B maximum value determining step, an RGB minimum value determining step of obtaining an RGB minimum value as a minimum value of the RGB data, an RGB median determining step of obtaining an RGB median value as a median of the RGB data, Median and R
A difference calculating step for obtaining a difference from the maximum GB value;
B color comparing step of comparing the median value with the threshold value, comparing the difference with the threshold value, and determining the color of the character image based on the RGB maximum value, the RGB minimum value, the RGB median value, the difference and the threshold value. Has the advantageous effect that the dependence of the threshold value is only the RGB median value and the difference, and the dependence of the threshold value can be reduced as compared with the related art.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of color determination of a yellow character in a color character determination method according to a first embodiment of the present invention;

FIG. 2 is an explanatory diagram of color determination of a blue character in the color character determination method according to the first embodiment of the present invention;

FIG. 3 is an explanatory diagram of color determination of a cyan character in the color character determination method according to the first embodiment of the present invention;

FIG. 4 is an explanatory diagram of color determination of a red character in the color character determination method according to the first embodiment of the present invention;

FIG. 5 is an explanatory diagram of color determination of a magenta character in the color character determination method according to the first embodiment of the present invention;

FIG. 6 is an explanatory diagram of color determination of a green character in the color character determination method according to the first embodiment of the present invention;

FIG. 7 is a flowchart showing a color character discrimination method according to the first embodiment of the present invention;

FIG. 8 is an explanatory diagram of a color character discrimination method according to the second embodiment of the present invention.

FIG. 9 is a flowchart showing a color character discrimination method according to the second embodiment of the present invention;

FIG. 10 is a flowchart showing a color character discrimination method according to the second embodiment of the present invention;

FIG. 11 is an explanatory diagram of a color character discrimination method according to a third embodiment of the present invention.

FIG. 12 is a flowchart illustrating a color character discrimination method according to a third embodiment of the present invention;

FIG. 13 is a flowchart showing a color character discrimination method according to the third embodiment of the present invention;

FIG. 14 is an explanatory diagram of a color character discrimination method according to a fourth embodiment of the present invention.

FIG. 15 is a flowchart illustrating a color character discrimination method according to a fourth embodiment of the present invention.

FIG. 16 is a flowchart illustrating a color character discrimination method according to a fourth embodiment of the present invention.

FIG. 17 is a flowchart showing a conventional color character discrimination method.

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁶ Identification code FI H04N 1/40 F

Claims (6)

[Claims] 1. A pixel determined to be a character image by image attribute determination processing of a character, a halftone dot, a photograph, etc.
A color character discriminating method for discriminating a color of a character image from a value of input data, comprising: a difference minimum value discriminating step of obtaining a difference minimum value as a minimum value of a difference between arbitrary two data of RGB data; An RGB minimum value determining step of obtaining an RGB minimum value as a minimum value among data values; and a color determining step of determining a color of a character image based on the RGB minimum value and the difference minimum value. Color character determination method. 2. A pixel determined as a character image by image attribute determination processing of a character, a halftone dot, a photograph, etc.
A color character discriminating method for discriminating a color of a character image from a value of input data, comprising: a difference minimum value discriminating step of obtaining a difference minimum value as a minimum value of a difference between arbitrary two data of RGB data; An RGB maximum value determining step of obtaining an RGB maximum value as a maximum value among data values; and a color determining step of determining a color of a character image based on the RGB maximum value and the difference minimum value. Color character determination method. 3. A pixel determined as a character image by image attribute determination processing of a character, a halftone dot, a photograph, etc.
A color character discrimination method for discriminating a color of a character image from a value of input data, wherein RG as a maximum value of RGB data
An RGB maximum value determining step for obtaining a B maximum value;
RG for finding RGB minimum value as minimum value of data
A B minimum value determining step, an RGB median value determining step of obtaining an RGB median value as a median value of the RGB data, comparing the RGB median value with a threshold value, and comparing the RGB maximum value and the RGB minimum value with each other. A color determining step of determining a color of a character image based on the RGB median value and the threshold value. 4. A method according to claim 3, further comprising the step of determining a threshold value from the RGB maximum value and the RGB minimum value between said RGB median value determination step and said color determination step. The described color character determination method. 5. A pixel determined to be a character image by image attribute determination processing for a character, a halftone dot, a photograph, etc.
A color character discrimination method for discriminating a color of a character image from a value of input data, wherein RG as a maximum value of RGB data
An RGB maximum value determining step for obtaining a B maximum value;
RG for finding RGB minimum value as minimum value of data
A B minimum value determining step, an RGB median value determining step of determining an RGB median value as a median value of the RGB data, a difference calculating step of determining a difference between the RGB median value and the RGB minimum value, Comparing the value with a threshold value, comparing the difference with the threshold value, and changing the color of the character image based on the RGB maximum value, the RGB minimum value, the RGB median value, the difference, and the threshold value. And a color discriminating step of discriminating. 6. Pixels determined as character images by image attribute determination processing of characters, halftone dots, photographs, etc.
A color character discrimination method for discriminating a color of a character image from a value of input data, wherein RG as a maximum value of RGB data
An RGB maximum value determining step for obtaining a B maximum value;
RG for finding RGB minimum value as minimum value of data
A B minimum value determination step; an RGB median value determination step of determining an RGB median value as a median value of the RGB data; a difference calculation step of determining a difference between the RGB median value and the RGB maximum value; Comparing the value with a threshold value, comparing the difference with the threshold value, and changing the color of the character image based on the RGB maximum value, the RGB minimum value, the RGB median value, the difference, and the threshold value. And a color discriminating step of discriminating.