WO2007110965A1

WO2007110965A1 - Color pattern extracting device and color pattern extracting program

Info

Publication number: WO2007110965A1
Application number: PCT/JP2006/306585
Authority: WO
Inventors: Misako Suwa; Katsuhito Fujimoto
Original assignee: Fujitsu Limited
Priority date: 2006-03-29
Filing date: 2006-03-29
Publication date: 2007-10-04
Also published as: JP4322310B2; JPWO2007110965A1

Abstract

[PROBLEMS] An object pattern can be correctly extracted with reference to a registered color even if the color of the object pattern is varied due to mix with the background color. [MEANS FOR SOLVING PROBLEMS] Installed in an HD (20) of a color pattern extracting device (10) is a color pattern extracting program (21) including a learning module (22) having a function of determining a correction parameter used to convert the pattern color acquired when a specific color pattern is displayed on an arbitrary background color into a color acquired when the specific color pattern is displayed on a basic background color by using two images showing the specific color pattern displayed on different background colors and an extracting module (23) for separating the inputted image into the pattern and the background, acquiring the colors of the pattern and the background, converting the object pattern color into a color estimated when the object pattern color is displayed on the basic background color by using these values and the correction parameter acquiring by the learning module (22), and extracting a pattern matching the basic pattern color from the converted object patterns.

Description

Specification

Color pattern extraction apparatus and color pattern extraction program

TECHNICAL FIELD [0001] The present invention relates to a color pattern extraction apparatus for extracting a pattern of a specific color with a color image power, and a color pattern extraction program for causing a computer to function as such an apparatus, and in particular, a stamp post-printed on color paper. It relates to a method suitable for extracting patterns such as.

Background art

[0002] Technology for extracting a pattern of a specific color has been conventionally used for applications such as extracting a pattern such as a stamp from a color form on which a pattern such as a stamp or a seal is post-printed. For example, a method is known in which the color of a post-printed pattern is registered, and the image pattern obtained by optically reading a form or the like is used to extract a color pattern that approximates the registered color. However, even if the color of the pattern formed by ink such as stamps is the same, if the background color, which is the color of the paper, is different, it may be mixed with the background color and become a different color. is there. If the color of the pattern in the image read in this way changes due to color mixing, the similarity with the color of the registered pattern is lost, and the pattern may not be extracted.

[0003] On the other hand, Patent Document 1 discloses a method of comparing a pattern color in an image with a registered color on the basis of a background in the image in order to remove the influence of color mixing with the background color. ing . This method calculates the angle Θc between the target pattern color vector in the image and the background color vector, and the angle Θt between the registered color vector and the background color vector in the image. When the difference between these angles is smaller than the predetermined threshold, it is determined that the color of the target pattern matches the registered color. By calculating the angle Θt for a plurality of registered colors, the color of the target pattern can be assigned to the registered color with the smallest angle difference, and the pattern corresponding to the specified color can be extracted.

Patent Document 1: Japanese Patent Laid-Open No. 10-104075

Disclosure of the invention Problems to be solved by the invention

However, in the method of Patent Document 1, the degree of color mixture between the pattern in the image and the background is large, and the color difference between the pattern color in the image and the registered color becomes large, the above angle difference is It becomes larger, and there is a higher possibility that the color judgment will be wrong and pattern extraction will fail.

[0006] In addition, in the method of Patent Document 1, when a pattern of a color having the same color system and different RGB values is mixed in the image, there is a possibility that a pattern of a color different from that specified by the registered color may be extracted. . For example, if there are stamps and characters of the same color but different RGB values on a chromatic area different from the basic background color at the time of registration, and you want to extract only the pattern of the post-printed stamp, it will be represented by ink. The stamp color is different from the stamp color on the basic background color registered by mixing with the background color. At this time, if the color of the printed character is close to the stamp color on the registered basic background color, the character pattern is erroneously extracted.

[0007] The present invention has been made in view of the above-described problems, and the problem is that the pattern is accurately determined based on the registered color even when the color of the target pattern changes greatly due to color mixing with the background. Another object of the present invention is to provide a color pattern extraction device that can be extracted in a simple manner and a color pattern extraction program that causes a computer to function as such a device.

Means for solving the problem

[0008] In order to solve the above-described problems, a color pattern extraction apparatus according to the present invention is an apparatus for extracting a pattern of a specific color with a color image power, and a basic background of a specific color pattern to be extracted. The combination of the basic pattern color and basic background color, which is the color on the color, and the reference pattern color and reference background color, which is a color on the reference background color different from the basic background color of the pattern of the specific color Based on at least two sets of color information, it is estimated that a pattern color obtained when a pattern of a specific color is displayed on an arbitrary background color is displayed on the basic background color. Learning means for obtaining a correction parameter for conversion to a target color, and separating the input image into a target pattern and a background to obtain a target pattern color and a background color, and obtaining these values and the learning means. Correction parameters Extraction means for converting the pattern color into a color estimated to be obtained when displayed on the basic background color, and extracting a pattern that matches the basic pattern color from the converted target pattern color. Features.

[0009] The learning means includes a learning pattern registration unit for registering a basic image in which a specific color pattern is represented on a basic background color and a reference image in which a specific color pattern is represented on a reference background color; For each of the image and the reference image, a separation unit that separates the pattern and the background, a color calculation unit that obtains a representative color of the separated pattern and the background, and a correction parameter is obtained based on the obtained representative color And a correction parameter setting unit.

[0010] In addition, the extraction unit includes a separation unit that separates an input image into a target pattern and a background, a color calculation unit that obtains a representative color of the separated target pattern and the background, and the obtained representative color A color mixing correction unit that converts the representative color of the target pattern based on the correction parameter; and a determination unit that determines whether the color of the target pattern after color conversion matches the basic pattern color. Is desirable.

[0011] As a color overlap model in the learning means and the extraction means, a particle density model is applied, and at least for the pattern color, a parameter is introduced into the density coefficient of the particle density model to determine the color for each pattern portion. It is desirable to reflect this difference in color conversion.

On the other hand, the color pattern extraction program according to the present invention allows a computer to set a combination of a basic pattern color and a basic background color, which is a color on a basic background color of a specific color pattern to be extracted, Based on at least two sets of color information of a reference pattern color and a reference background color, which is a color on the reference background color different from the basic background color of the color pattern, the pattern of the specific color is arbitrarily set A learning function for obtaining correction parameters for converting the pattern color obtained when displayed on the background color into a color estimated to be obtained when displayed on the basic background color, and the input image When the target pattern color and background color are obtained separately from the target pattern and background, and the target pattern color is displayed on the basic background color using these values and the correction parameters obtained by the learning means Presumed to be obtained It is converted to the color being characterized in that to execute an extracting function of extracting a pattern matching the basic pattern color from converted by target pattern.

The invention's effect

[0013] According to the color pattern extraction apparatus of the present invention configured as described above, the background color is added. Even if the color of the post-printed pattern changes significantly due to a color mixture with the background color, the pattern color can be corrected to the color displayed on the basic background color, and the matching with the specified color can be determined. Can be accurately performed to extract a color pattern.

BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, embodiments of a color pattern extraction apparatus and a color pattern extraction program according to the present invention will be described with reference to the drawings. First, the outline of the entire system including the color pattern extraction apparatus of this embodiment will be described based on FIG. As shown in FIG. 1, the system 1 of the embodiment includes a color pattern extraction apparatus 10 and a display 30, a printer 31, and a scanner 32 connected to the color pattern extraction apparatus 10.

The color pattern extraction apparatus 10 is configured by a general-purpose computer such as a personal computer, and includes a CPU 11, a hard disk (HD) 20, a memory (RAM) 12, and an interface 13 connected to the CPU 11. An operating system (not shown) is installed in the HD 20 and a color pattern extraction program 21 is installed. The display 30, printer 31, and scanner 32 are connected to the CPU 11 via the interface 13. The CPU 11 expands and executes the color pattern extraction program 21 on the RAMI 2 as necessary. The target color image is read by the scanner 32 or provided as an image file. The extraction process and results are displayed on the display 30 as needed and printed out by the printer 31.

[0016] The color pattern extraction program 21 includes a learning module 22 that registers a pattern or background color as a pre-processing for extraction and obtains a correction parameter, and an extraction module 23 that extracts a pattern of actual image power.

[0017] The learning module (learning means) 22 is a combination of a basic pattern color and a basic background color, which is a color on a basic background color of a pattern of a specific color to be extracted, and a basic pattern of the specific color. When a pattern of a specific color is displayed on any background color based on at least two sets of color information of a reference pattern color and a reference background color, which is a color on the reference background color different from the background color It has a function to obtain correction parameters for converting the obtained pattern color into a color estimated to be obtained when displayed on the basic background color. And learning module 22 includes a learning pattern registration unit 22a for registering a basic image in which a specific color pattern is represented on a basic background color and a reference image in which a specific color pattern is represented on a reference background color, and a basic image For each of the reference image and the reference image, a separation unit 22b that separates the pattern and the background, a color calculation unit 22c that calculates a representative color of the separated pattern and the background, and a correction parameter based on the obtained representative color And a correction parameter setting unit 22d. A representative color is a color assigned to each pattern by, for example, averaging the colors of pixels constituting the pattern. Each color is calculated by taking an average value after extracting a pattern.

On the other hand, the extraction module (extraction means) 23 obtains the target pattern color and the background color by separating the input image into the target pattern and the background, and these values and the correction obtained by the learning module 22 Using the parameters, the target pattern color is converted to the color estimated to be obtained when displayed on the basic background color, and the converted target pattern force has a function to extract a pattern that matches the basic pattern color . The extraction module 23 includes a separation unit 23a that separates the input image into a target pattern and a background, a connected component calculation unit 23b that separates the target pattern for each connected component, and representatives of the background color and each connected component. The color calculation unit 23c for obtaining the color, the color mixing correction unit 23d for converting the representative color of the target pattern based on the obtained representative color and the correction parameter, and the color of the target pattern after the color conversion match the basic pattern color And a determination unit 23e for determining whether or not the force to perform is included. The determination unit 23e determines a match between the corrected target pattern and the designated basic pattern color using a known method. For example, the case where the angle between the color vector of the target pattern after correction and the color vector of the basic pattern color is equal to or smaller than the threshold value is matched, and the case where the angle exceeds the threshold value is determined to be mismatched.

Subsequently, the extraction principle of the color pattern extraction apparatus of the present invention will be described based on a specific example. In the following description, an image obtained by reading a paper on which a handwritten pattern with ink is post-printed by the scanner 32 is a processing target, a paper color is a background color, and a post-printed handwritten pattern is an extraction target. The basic background color is white.

[0020] As shown in FIG. 2 (A), an object that is post-printed with a specific color ink on the basic background color S

0

Let C be the representative color of the pattern (handwritten A). C corresponds to the basic pattern color. Also,

A A

As shown in Fig. 2 (B), the same color as above with the reference background color S that is different from the basic background color. C represents the representative color of the target pattern (handwritten ABCD) post-printed with a specific color ink

A

'And. C ′ corresponds to the reference pattern color. In Fig. 2 (B), it is extracted in the same image.

A

The pattern “item name” of the representative color D is printed.

The learning module 22 obtains correction parameters based on the two sets of color information (SO, CA) and (SI, CA,). In this example, a known particle density moire (Terai et al., Base Recognition by etragonal Regression for Ovena pped Watercolors ", Proc. Of lWFHR-9, 2004) is used as a color mixture model for defining color overlap. Using this model, the colors C and C 'of the target pattern post-printed on the background colors S and S are respectively

0 1 A A

It is expressed as equations (1) and (2). Since S is the basic background color (white), C 'is as shown in equation (3)

0 A

In addition, C and S are mixed colors.

A 1

[0022] [Equation 1] ^ ni / ^ + hiichi ^. (1)

C, =, T, + (1- ^) 8, (2)

C ^ C.--^ X ^ -S ,,) (3) [0023] where T is the ink color (specific color) of the target pattern and is the ink density of the target pattern.

A A

Corresponding coefficient, 0.0≤ p ≤1 if the color is uniform for the target color surface

A

It can be thought of as a constant with a value of .0. As described above, if two sets of background color and pattern color are known, p can be obtained. And if p is determined, on any background color

A A

It is possible to correct the color of the pattern that has been mixed by post-printing with a specific color of ink and convert it to a color estimated to be obtained when displayed on the basic background color (mixed color correction). A is generally determined by the color distribution (not the representative color of the pattern) of each pixel composing the pattern.

For example, when color mixture correction is performed using the image in FIG. 2 (B) as the target image, the pattern colors C,, and D are used as the basis.

A

The background color S is converted to the color C '+, D +, which is presumed to be obtained when displayed on S. Figure 2 (C)

0 A

An image as shown in FIG. Fig. 3 (A) is a vector indicating the background color and pattern color of the image of Fig. 2 (B), and Fig. 3 (B) is the background color and pattern of the image of Fig. 2 (C) after color mixture correction. This is a vector indicating the representative color.

Here, FIG. 3 (A) shows the non-target pattern “item” for the basic pattern color C color vector. Name “Color” Angle formed by D color vector 0 Force Basic pattern color

1 A

It is assumed that the color vector of the elephant pattern (handwritten ABCD) is smaller than the angle Θ.

2

The In this case, if the pattern is to be extracted without correcting the color mixture in the state shown in Fig. 3 (A) as in the conventional case, if the threshold value for judging that the color is the same is set to 0, the pattern is not shown as shown in Fig. 4 (A). versus

1

Only the elephant pattern is extracted, the pattern that should be extracted is dropped, and the threshold is set to Θ

2 and non-target patterns that are not only target patterns as shown in Fig. 4 (B) are also extracted.

On the other hand, when color mixture correction is performed, the color of the target pattern changes from C ′ to C as shown in FIG. 3 (B).

A

It is converted to '+ and approaches the basic pattern color C of the specific color, and the color of the non-target pattern is changed from D

A A

Converted to D + and moved away from basic pattern color C. As a result, the basic pattern color C

A A

The angle Θ + formed by the color vector of the color D + after correcting the non-target pattern with respect to the vector is

From angle Θ + formed by the color vector of the target pattern with respect to the color vector of one pattern color C

A 2 Large. For this reason, if the threshold value for judging the same color is 0+, as shown in Fig. 4 (C)

2

Only the target pattern is extracted.

[0027] When the target pattern is a line drawing such as a character, the color distribution of each pixel constituting the line pattern has extremely high dispersion in one axial direction, and becomes a distribution form on the line. The assumption that, is a constant is difficult to hold. However, the background is almost the same as the surface figure.

A

It can be assumed to be uniform. An example of the color distribution for each pixel when the pattern is a line drawing will be described with reference to FIG. Figure 5 shows the color distribution E of each pixel that makes up the pattern printed on the basic background color, and each pixel that makes up the pattern printed with the same ink on a predetermined background color different from the basic color. Color distribution F, color distribution G of each background pixel of the basic background color, and color distribution H of each background pixel of the predetermined background color. Color correction means conversion from F to G.

[0028] Therefore, in the embodiment, a parameter t representing the fluctuation of is introduced and replaced as shown in Equation (4).

A

Yeah. The range of t is as shown in Equation (5).

[0029] [Equation 2]

0 <ί <1 (5) [0030] Further, it is assumed that (0) = 0 and P (1) = 1. Therefore, equations (1) to (3) can be rewritten as the following equations (6) to (8).

A

available.

[0031] [Equation 3]

C _A (t) = p _A (tyi _A + \ -p _A (t)) S,

Cri = (ri- (1-) (-s ₀ )

[0032] From Equations (6) and (7), Equation (10) 11) holds true at t = 1 and t = 0, respectively.

[0033] [Equation 4]

C, (1) = C, (1) = T, (10)

C, (0) = SC ^ (0) = S ₀ (11)

[0034] Therefore, two approximate straight lines of C (t) -S and C '(t) -S are obtained from the distribution, and the two straight lines

A O A 1

Τ can be obtained by obtaining the intersection coordinates. C for a pixel

A A 'ω is observed, and t = t at this time. The following equation (12) is derived from equation (7). On the other hand, the background color

P

Approximation as shown in Eq. (13) using the mean value of the distribution.

[0035] [Equation 5] c (-

(12)

I -s' l—

(13)

T _A -S _X

However, S s. Is the distribution average.

[0036] Equation (14) is derived from the above equation and equation (8). Using this equation, this is displayed on the basic background color based on the pattern color C, (t) of the input image. Color C (t)

A A

wear.

[0037] [Equation 6]

C tX (t) + ri ^ ¹¹ ) (^ "^) (14) Based on the above derivation, two combinations of the background color and the pattern color on the background color are prepared. It can be seen that the pattern color of can be restored. Module 22 learns the above T from two sets of colors.

A

[0039] It should be noted that the particle density model of Terai et al.

A

Differences from the extended particle density model of the present application, in which data is introduced, will be explained conceptually with reference to Figs. Assuming that the color distribution of the background pixels is almost uniform in all models,

0 1 shows. On the other hand, the color distribution of each pixel of the pattern post-printed on the background color is assumed to be almost uniform by Terai et al.

A A

In contrast, the extended particle density model of the present application introduces a parameter to

A

It is assumed that the distribution is on a straight line in consideration of the distribution.

[0040] When trying to convert the color distribution of a line drawing pattern that was post-printed on a background color different from the basic background color in Terai's model to a color on the basic background color, as shown in Figure 8 It is not converted to a correct distribution. Therefore, the line drawing pattern may not be extracted correctly. On the other hand, according to the model of the present application, C (t) -S, C and (t) -S

A O A 1

By calculating the line, the color distribution of the line drawing pattern that has been post-printed on a background color different from the basic background color is correctly converted to the color distribution that is expected to be obtained when displayed on the basic background color. Can do. Therefore, when extracting line drawing patterns such as characters, the extraction accuracy can be improved by using the extended particle density model of the present application.

[0041] Next, the processing flow of the color pattern extraction program 21 of the embodiment will be described based on a flow chart. When the processing of the learning module 22 is started, a learning image pattern is registered by the action of the learning pattern registration unit 22a as shown in FIG. 9 (S01). Here, the image of the target pattern post-printed on the basic background color to be extracted and the image of the target pattern post-printed with the same ink on a reference background color different from this basic background color are registered. . FIG. 10 (A) shows an example of an image to be registered.

[0042] Next, by the operation of the separation unit 22b, each of the two registered images is binarized by an existing method and separated into a no-turn and a background (S02). Figure 10 (B) shows a binary binary image with a separated background force. The color calculation unit 22c then displays the representative color of the basic background color and the color of each pixel, the representative color of the reference background color and the color of each pixel, the representative color of the target pattern post-printed on each background color, and each pixel. Calculate the color (S03). FIG. 10 (C) shows an example in which the background and the pattern are displayed in respective representative colors for each image. [0043] The correction parameter setting unit 22d calculates parameters in the color mixture correction formula of the target pattern in accordance with the above extended particle density model (S04). The representative color of the target pattern on the basic background color registered in the learning pattern registration unit 22a is registered as a color that can be selected when the user specifies it during pattern extraction. The basic background color and the correction parameters are Are registered in the parameter file as coefficients used in

[0044] When the process of the extraction module 23 starts, the user selects and designates a pattern color (extracted ink color) from which the color power registered by the process of the learning module 22 is to be extracted (Sll). Subsequently, an extraction target image in which a pattern is post-printed with the same ink on an arbitrary background color is input (S12). An example of the input image is shown in Fig. 12 (A). The separation unit 23a separates the image input by the binary key into a pattern and a background (S13). The binary image of the separated pattern is shown in Fig. 12 (B).

[0045] The connected component calculation unit 23b performs labeling or the like on the binary image, and extracts a noturn force connected component (S14). Fig. 12 (C) shows the calculation result of the connected component, and the part enclosed by the dotted circumscribed rectangle is one connected component. Subsequently, the color calculation unit 23c calculates a representative color of the background color and a representative color of each connected component of the pattern (S15). Figure 12 (D) shows the representative colors for each connected component.

[0046] The color mixing correction unit 23d uses the correction formula obtained by the learning module 22 to perform color mixing correction using the background color and the representative color of the connected component for each connected component, and displays it on the basic background color. The case color is estimated (S16). FIG. 12 (E) shows the corrected pattern. The judgment unit 23e compares the representative color after the color mixture correction of each connected component with the specified color (representative color of the pattern for which extraction is specified) (S17), and determines whether the corrected pattern color matches the specified color. As shown in FIG. 13, it is determined for each connected component by threshold processing (S18). When it is determined that they match, the target connected component is extracted as a target pattern (S19), and when it is determined that they do not match, the target connected component is deleted (S20). If you want to make a determination in units smaller than the connected components, for example, in units of pixels, you can perform color mixture correction and determine the match with the specified color in units of pixels without calculating the connected components in S14.

The extraction result is displayed on the display 30 or printed out by the printer 31 according to the user's request. At that time, as shown in Fig. (Shown on a single scale), or as a binary image as shown in FIG. 14 (B).

Brief Description of Drawings

FIG. 1 is a block diagram showing a system including a color pattern extraction apparatus according to an embodiment of the present invention.

[Fig. 2] (A) is an image of a basic background color post-printed with a specific color ink, (B) is an image of a reference background color with a specific color ink post-printed, (C) is an image FIG. 5B is an explanatory diagram showing an image obtained by correcting color mixture with the apparatus of the embodiment (B).

3A is an explanatory diagram showing a color vector of the image of FIG. 2B, and FIG. 3B is an explanatory diagram showing a color vector of the image of FIG. 2C.

[FIG. 4] (A) and (B) are explanatory diagrams showing extraction results by a conventional extraction method, and (C) are explanatory diagrams showing extraction results by the apparatus of the embodiment.

FIG. 5 is a graph showing an example of color distribution for each pixel when the pattern is a line drawing.

FIG. 6 is an explanatory diagram conceptually showing color mixture correction by a particle density model of Terai et al.

FIG. 7 is an explanatory diagram conceptually showing color mixture correction by an extended particle density model of the embodiment.

FIG. 8 is an explanatory diagram conceptually showing the result of color mixture correction of the color distribution of a line drawing pattern using a particle density model of Terai et al.

FIG. 9 is a flowchart showing the processing contents of a learning module of the color pattern extraction apparatus of the embodiment.

[Fig. 10] Examples of images processed in Fig. 9. (A) is a registered image, (B) is a binary image separated from the background, and (C) is a representative of the background and pattern. It is explanatory drawing which shows the image displayed with the color.

FIG. 11 is a flowchart showing the processing contents of the extraction module of the color pattern extraction apparatus of the embodiment.

FIG. 12 is an example of the image processed in FIG. 11, where (A) is the input image, (B) is the binary image of the separated pattern, (C) is the calculation result of the connected component, and (D) is Representative colors for each connected component, (E) is an explanatory diagram showing a pattern after color mixture correction. FIG. 13 is an explanatory diagram showing a state of threshold processing for each connected component in the processing of FIG. 11.

[FIG. 14] (A) is an image showing a color image of the extraction result in gray scale, and (B) is an explanatory diagram showing a binary image of the extraction result.

Explanation of symbols

10 Color pattern extraction device

11 CPU

13 RAM

14 interface

20 HD

21 Color pattern extraction program

22 Learning Department

23 Extractor

Claims

The scope of the claims

[1] Color image power A color pattern extraction apparatus for extracting a pattern of a specific color, which is a set of a basic pattern color that is a color on a basic background color of the pattern of the specific color to be extracted and the basic background color And the specific color based on at least two sets of color information of a reference pattern color and a reference background color which are different from the basic background color of the pattern of the specific color Learning means for obtaining a correction parameter for converting a pattern color obtained when the pattern is displayed on an arbitrary background color into a color estimated to be obtained when the pattern color is displayed on the basic background color;

The input image is separated into the target pattern and the background to obtain the target pattern color and the background color, and using these values and the correction parameter obtained by the learning means, the target pattern color is determined as the basic pattern color. An extraction means for converting to a color estimated to be obtained when displayed on a background color, and extracting a pattern whose converted target pattern force also matches the basic pattern color;

A color pattern extraction apparatus comprising:

[2] The learning means includes

A learning pattern registration unit for registering a basic image in which the pattern of the specific color is represented on the basic background color and a reference image in which the pattern of the specific color is represented on the reference background.

A separation unit for separating the basic image and the reference image into a pattern and a background;

A color calculation unit for obtaining a representative color of the separated pattern and the background;

The color pattern extraction apparatus according to claim 1, further comprising: a correction parameter setting unit that obtains the correction parameter based on the obtained representative color.

[3] The extraction means includes

A separation unit that separates an input image into a target pattern and a background;

A color calculator for obtaining representative colors of the separated target pattern and the background;

Based on the obtained representative color and the correction parameter, the representative color of the target pattern A color mixing correction unit for converting

A determination unit for determining whether or not the color of the target pattern subjected to color conversion matches the basic pattern color;

The color pattern extraction apparatus according to claim 1, further comprising:

[4] A particle density model is applied as a color overlap model in the learning unit and the extraction unit, and at least for the pattern color, a parameter is introduced into the density coefficient of the particle density model to obtain a portion of the pattern. 4. The color pattern extraction apparatus according to claim 1, wherein a difference in color for each color is reflected in color conversion.

[5] A color pattern extraction program that makes a computer yesterday as a color pattern extraction device that extracts a pattern of a specific color with color image power,

On the computer,

A set of a basic pattern color that is a color on the basic background color of the pattern of the specific color to be extracted and the basic background color, and a reference background color that is different from the basic background color of the specific color pattern A pattern color obtained when a pattern of the specific color is displayed on an arbitrary background color based on at least two sets of color information of a reference pattern color that is a color of the reference color and the reference background color. A learning function for obtaining a correction parameter for conversion to a color estimated to be obtained when displayed on the background color;

The input image is separated into the target pattern and the background to obtain the target pattern color and the background color, and using these values and the correction parameter obtained by the learning means, the target pattern color is determined as the basic pattern color. An extraction function for converting to a color estimated to be obtained when displayed on the background color, and extracting a pattern whose converted target pattern power also matches the basic pattern color;

A color pattern extraction program characterized in that