JPH10145616A - Color converter and color conversion method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent coloring in an achromatic character and a gray scale image and to suppress a sense of roughness in a natural image in a print image based on color conversion processing. SOLUTION: Based on picture element values (R, G, B) of a specified picture element in the color conversion processing, at first an achromatic color discrimination means 101 discriminates whether or not a color of the picture element is achromatic. When the color is discriminated not to be achromatic, the picture element is processed as a natural image, picture when the color is discriminated to be achromatic, the picture element is given to a natural image/text image discrimination means 102, where the image is a natural image or a text image is discriminated. Thus, the picture element value (Y, M, C) of the picture element discriminated to be a natural image or a text image is is selected by a selector 103, from which the selected picture element value is transferred to a natural image processing interpolation arithmetic means 104 whose lookup table(LUT) is optimized for the natural image or a text image processing interpolation arithmetic means 105 whose lookup table(LUT) is optimized for the text image, where each conversion processing is conducted.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color conversion device and a color conversion method for performing color conversion according to the type of an output image.

[0002]

2. Description of the Related Art FIG. 1 shows a conventional example of this type of color conversion apparatus. The configuration shown in the figure uses a lookup table (hereinafter, also referred to as LUT) and an interpolation operation together.

That is, an intermediate density signal (Y, M, C) obtained by converting a luminance signal (R, G, B) to approximate a density by a luminance-density conversion unit 1 is used by an interpolation calculating means 2. , The color material control signals (Y _P , M _P ,
C _P , K _P ). Then, the printer 3 generates a print output 4 based on the color material control signal. Here, the interpolation calculation means 2 receives a three-dimensional input (Y, M, C).
, And a three-dimensional LUT for outputting a color material control signal on the basis of the above.

Various methods of interpolation calculation using the output of such a three-dimensional LUT are known, for example, Japanese Patent Application Laid-Open No. 63-162248 and Japanese Patent Publication No. 58-16180.
And Japanese Patent Application Laid-Open No. 3-229573 are known.

[0005]

However, in the above-described conventional configuration, the following problem may occur.

In general, an LUT is often used after being optimized according to an image to be printed.
Is set to be optimal for outputting a natural image, the output amount of black in the low-density part to the medium-density part is set relatively low in the LUT. For this reason, for example, when printing achromatic characters (when Y = M = C), if there is a so-called misregistration in the printer, since the amount of black is small, the misregistration portion will appear colored. Sometimes. Further, for the same reason, an achromatic image generated by computer graphics (hereinafter abbreviated as CG) or the like may be colored.

Conversely, if the LUT is set to increase the amount of black relatively in accordance with the text image, the black mixes with the memory color such as the flesh color existing in the natural image. May cause a feeling of roughness.

The present invention has been made in view of the above problems, and has as its object to prevent coloring of achromatic characters and grayscale images to be printed, and to reduce the roughness of natural images. It is an object of the present invention to provide a color conversion device and a color conversion method that can suppress the occurrence of color.

[0009]

For this purpose, the present invention provides a color conversion apparatus for performing color conversion using a look-up table, wherein a determination is made as to whether an image to be printed is a natural image or a non-natural image. Means and a data conversion process using a look-up table suitable for a natural image, or another data conversion process using a look-up table suitable for an image other than the natural image, based on the determination by the determination unit. Data conversion means for performing either of the above.

[0010] A color conversion method for performing color conversion using a look-up table is provided, and it is determined whether an image to be printed is a natural image or an image other than the natural image. Performing a data conversion process using a look-up table suitable for an image or another data conversion process using a look-up table suitable for an image other than the natural image. And

According to the above arrangement, the image to be output is
The determination unit determines whether the image is a natural image or a text image other than the image. Depending on the determination, a conversion process using a look-up table adapted to the natural image or a conversion process using a look-up table adapted to the text image is performed. Therefore, for example, in a look-up table for a natural image, it is possible to set a process of relatively lowering the output amount of black in a low-density part to a medium-density part, while, in a look-up table for a text image, , A conversion process that increases the amount of black can be set.

[0012]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 2 is a block diagram showing the configuration of the color conversion apparatus according to one embodiment of the present invention. In the figure, elements having the same functions as the elements shown in FIG. 1 are denoted by the same reference numerals, and detailed description thereof will be omitted.

In FIG. 2, reference numeral 107 denotes a line buffer, for example, the number of pixels to be printed in one scan of the print head in the printer 3, the number of columns × n, the density values R, G, B (hereinafter, pixel values) ) Is two-dimensionally expanded and stored. Note that this line buffer 1
07 is provided corresponding to each of the R, G, and B surfaces.

Reference numeral 101 denotes an achromatic color judging means for judging whether or not the pixel of interest located at the center of the window of size n × n defined in the line buffer 107 is achromatic as described later. . Reference numeral 102 denotes a natural image / text image determining unit, which determines whether or not the pixel is included in the natural image region when the target pixel is achromatic, as described later. In the following description, the “text image” refers to an image including an achromatic character and an achromatic grayscale image found in a CG image.

Reference numeral 103 denotes a selector, which outputs an intermediate density output from the luminance-density conversion unit 1 in accordance with a result of the determination by the achromatic color determining means 101 or the natural image / text image determining means 102, by a natural image processing interpolation calculating means. 104 or the text image processing interpolation calculation means 105.

The natural image processing interpolation operation means 104 comprises a three-dimensional LUT and an interpolation operation means, and the three-dimensional LUT is optimized for natural image processing. Note that various methods can be used as the interpolation calculation method, and for example, those disclosed in the respective publications described with reference to FIG. 1 can be used. on the other hand,
The text image processing interpolation calculation means 105 is composed of a one-dimensional LUT optimized for processing an achromatic text image and an interpolation calculation means. The reason why the one-dimensional LUT is used is that when the color represented by the signal (Y, M, C) is an achromatic color, it is assumed that Y = M = C. In this sense, the luminance-density conversion unit 1
Requires that when R = G = B, a conversion such that Y = M = C is performed. In addition, as a method of the interpolation calculation in the means 105, in addition to a generally known linear calculation method using two points on a straight line, a non-linear calculation method using more points can be applied.

It is well known that the natural image processing interpolation calculation means 104 and the text image processing interpolation calculation means can be constituted only by the LUT without using the interpolation calculation.

The control means 108 controls the following elements:
While exchanging data with the host device 201, FIG.
The color conversion processing is executed by controlling according to a procedure described later.

Hereinafter, processing based on the configuration of the color conversion apparatus shown in FIG. 2 will be described mainly with reference to a flowchart shown in FIG. In the following, the line buffer 107 has a capacity of 5 rows for simplification of the description.
Although the processing window defined in the line buffer 107 is treated as having a 5 × 5 area, this is merely for the purpose of simplifying the description, and the application of the present invention is limited to these sizes. Obviously, there is no need for the window area and the number of lines in the line buffer to be odd numbers equal to each other.

The configuration for the color conversion processing shown in FIG. 2 below can be configured as a hardware circuit including a CPU and a memory, and can also be configured as software based on the control of the CPU. . Further, the printer 3 may be an independent printer or a printing unit (printing device) in a copying machine, and the host device 201 may be a personal computer, for example.

FIG. 3 is a flowchart showing the procedure of the color conversion process based on the configuration shown in FIG.

First, in step S101, a row number holding buffer W _i : (where i = 0,..., 4) is initialized. As shown in FIG. 4, this buffer stores data for associating the Y coordinate in the coordinate system defined inside the processing window 401 with the line numbers assigned to the five line buffers 107. Is what you do. That is, in FIG. 4, 0 to 0 along the X axis and the Y axis.
Although the numeral 4 is assigned, this range represents the processing window 401. A coordinate system representing a pixel position in the entire image area 403 is shown as an (X ′, Y ′) coordinate system. Moreover, although the line numbers in the line buffer 107 is swung from 0 to 4, which stores pixel values of lines represented by the coordinate Y '= l _j in the image area 403 in the row number k of the line buffer 107 Indicates that it is being done. For example, in the example of the figure, the pixel value of l ₁ line is stored in the buffer represented by the line number 3. In addition, line buffer 1
07 has an extra storage area for two pixels corresponding to the left and right ends of the image area, which is necessary to cover the existing area at the left and right ends according to the movement of the window 401. However, no actual pixel value is stored in this portion.

The relationship between the row number k and the coordinates Y is given by W _Y = k. That is, in the example of FIG.

[0025]

(Equation 1)

This is the result of the line buffer 10
In FIG. 7, the row of row number k storing the pixel values of l _j rows of the pixel area 403 means that it corresponds to the Y coordinate Y of the window 401.

Referring again to FIG. 3, step S10
In the initialization at 1, W _i = i: (i = 0,..., 4)
The processing is performed so that Further, in step S102, the line buffer 107 is initialized. In this initialization process, a value (for example, −1) that cannot be input as an input value is written to the line buffer 107. Next, at step S103, reads the pixel value of the first row of the image region 403 in the line buffer 107 of the line number stored in the row number holding buffer W _2. Thus, this first row corresponds to Y = 2 in the processing window 401. Similarly, read a second row of the image region 403 corresponds to the line number holding buffer W _3. Further, in step S104, the image area 403
Then, C0 = 3 is set for 0 by the counter indicating the row, and all the initialization work is completed. FIG. 5 shows a state at the time when the process of step S104 is completed. From the figure, it is understood that the ready to start the process by reading the pixel value corresponding to the line number holding buffer W ₄ is in place.

[0028] At next step S105, the value of the counter C0 line number holding buffer W _4, i.e., read 3. In step S106, the window 401 is moved according to the value of the counter retu. That is, the value of the counter retu starts from 1 and increases to a value indicating the right end in the row direction of the image. This means that the window 401 in FIG. 4, that is, the pixel of interest therein moves from the left end of the image to the right end of the image.

In step S107, the window 401
It is determined whether or not the pixel of interest at the center of is achromatic. This determination condition is that the pixel value of the pixel of interest represented by the coordinates (X, Y) in the window coordinate system is (R (X,
Y), G (X, Y), B (X, Y))

[0030]

(Equation 2)

Is given by

If the result in step S107 is False, the flow directly proceeds to the natural image processing in step S110. On the other hand, when the determination result is True,
The process proceeds to a natural image / text image determination process in step S108. The processing in step S108 is performed with reference to FIG.

First, in step S201, the number of pixels having a pixel value equal to the pixel value of the pixel of interest is determined in window 40.
1. Count inside. Then, in step S202, the number of pixels is compared with a threshold Th1. If the number of pixels is larger than Th1, it is determined that the image is likely to be a text image, and the process proceeds to step S203. If the number of pixels is T
If h1 is smaller than h1, it is determined that the image is a natural image, and the flow branches to step S211.

This determination is based on the statistical characteristics of the image. FIG. 7 shows a frequency distribution for explaining this feature. In the figure, the horizontal axis represents the number of pixels obtained in step S201, and the vertical axis represents the normalized frequency obtained by dividing the frequency of members belonging to the horizontal axis by the total number of detected achromatic pixels. That is, the result in step S201 (the number of achromatic colors having the same value) is i (i = 1,..., 25).
, The normalization frequency Pi
Is

[0035]

(Equation 3)

Is represented by Here, (a) represents the frequency distribution of a typical natural image, and (b) represents the frequency distribution of a document image composed of black characters of 6 pt to 12 pt. Note that these characteristics schematically represent experimental data and do not quantitatively represent the data itself.

As can be seen from the figure, the natural image has one pixel.
, Has a one-sided normal distribution property having a local maximum, and the document image exhibits multimodal characteristics due to a difference in character size. If the threshold value Th1 is set to be smaller than the number of pixels at which the characteristic shown in (b) saturates to 0, when the number of pixels is smaller than that, it can be determined that the image is a natural image with almost 100% probability. Here, the set value of the threshold value Th1 for setting the false detection rate (the probability that a character is determined to be a natural image) to 0 is determined by the size of the target character. For example, if the threshold value Th1 is set at the position A in the drawing, an erroneous detection rate of almost 0% can be achieved for characters of 6 pt or more. If the threshold value Th1 is set at the position B in the drawing, the same can be said for characters of 10 pt or more. That is, the threshold value Th1 is a parameter for adjusting the sensitivity.
Instead of lowering the character detection sensitivity, the probability of shifting to the processing after step 203 in the natural image area is reduced, and the processing efficiency is improved. Conversely, if the threshold value Th1 is reduced, the character detection sensitivity increases, but the processing efficiency decreases. Therefore, it can be said that the processing in steps S201 and S202 adjusts the trade-off relationship between the processing efficiency and sensitivity of the entire “natural image / text image determining unit”.

Referring again to FIG. 6, step S20
In step 3, the average value of the values of all pixels having a value different from the pixel value of the pixel of interest in the window is obtained.

[0039]

[Outside 1]

The pixel values of interest (R (2,2), G (2,2)
2), converts B (2,2)) of the value of a color difference system, respectively _{(Y ', C r',} C b '), put the (Y, 0,0). Here, the conversion from the (R, G, B) system to the color difference system is generally given as follows.

[0041]

(Equation 4)

Although there are various coefficients to be used depending on the definition of the color difference system, any coefficient may be used as long as the above condition is satisfied.

Next, in step S205, step S2
The absolute value of the difference between the two sets of pixel values in the color difference system obtained in step 04 is obtained as follows.

[0044]

Equation 5] Y "= | Y'-Y | C r" = | C r '| C b "= | C b' | In step S206, (Y" _{of, C r ", C b"} ), When at least one element is larger than Th2, it is determined that the pixel belongs to the character area, and it is determined that the pixel is a text image in step S210. In general, even if achromatic characters exist in a natural image, the purpose of the image creator is to make them stand out even if they exist in a natural image. There is great contrast. In steps S203 to S206, the character portion is separated using this property. Here, the same threshold value Th2 is used for the three components, but it is a matter of course that different threshold values may be used for these components.

Next, in steps S207 to S209, an achromatic grayscale image created by CG or the like is determined. This image can be displayed vertically or horizontally,
This is an image in which the gradation of gray gradually changes. Therefore, it cannot be determined by the method based on the background and the edge as used in steps S203 to S206. here,
A grayscale image is determined using binarization and pattern matching. First, in step S207, all pixels having a value equal to the pixel value of the target pixel in the window 401 are set to 1 and all other pixels are set to 0. Then, in step S208, the pattern to be processed is compared with each of the plurality of patterns shown in FIG. 8, and if any one of the patterns matches, it is determined in step S209 that the pattern is a gray scale. In step S210, the image is determined to be a text image.

As described above, when the processing in step S108 is completed, in step S109 shown in FIG. 3, if the image is a natural image, the processing in step S110 is performed based on the result determined in step S108. If it is a text image, the flow branches to perform processing in step S111.

Thus, the process for one pixel is completed, and the process returns to step S106, where the process for one row is sequentially performed.

When the processing for one line is completed, step S1
Updates the contents of the line number holding buffer W _i as follows: in 12.

[0049]

W _i = (W _i +1) mod 5 As a result, the window 401 moves the image area 403 downward by one line.

In step S113, the counter C0 is checked. If the last line has already been read, the flow branches to step S115, which is line ending processing. Otherwise, the counter C0 is incremented and step S105 is executed.
Return to the processing of the next line.

Finally, the operation in step S115 is performed with reference to FIG.

FIG. 10 shows the state at the time of branching to step S115. Image does not exist to be read no longer in the corresponding line buffer to the line number holding buffer W ₄ As is clear from FIG. It can also be seen that two unprocessed rows remain. Therefore, the loop counter in step S301 is set to perform processing for two lines, and in step S302, initialization is performed instead of reading an image in accordance with the value of the buffer W4. This initialization is performed in the same manner as in step S101 shown in FIG.
This is performed by substituting 1 or the like. Subsequent processing is the same as the processing shown in FIG. 3, and a description thereof will be omitted.

[0053]

As described above, according to the present invention, an image to be output is determined to be a natural image or another text image, for example, by a determination means, and the image adapted to the natural image is determined according to the determination. A conversion process using a look-up table or a conversion process using a look-up table adapted to a text image can be performed. For example, in a look-up table for a natural image, the appearance of black in a low density portion to a medium density portion is reduced. A process for relatively lowering the power can be set, while a conversion process for increasing the amount of black can be set in the look-up table for a text image.

As a result, it is possible to prevent coloring in a gray-scale image included in an achromatic character or a CG image, and to prevent graininess in a memory color such as a flesh color.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a conventional color conversion device.

FIG. 2 is a block diagram illustrating a configuration of a color conversion apparatus according to an embodiment of the present invention.

FIG. 3 is a flowchart illustrating a procedure of a color conversion process according to the embodiment.

FIG. 4 is a diagram for explaining a process for each pixel for determining a natural image / text image in the color conversion process.

FIG. 5 is a diagram for explaining processing of a first pixel in the determination processing;

FIG. 6 is a flowchart showing a procedure of a natural image / text image discriminating process.

FIG. 7 is a diagram for explaining setting of a threshold value for determining a natural image in the determination process.

FIG. 8 is a diagram showing a pattern for determining a text image in the determination processing.

FIG. 9 is a flowchart showing details of a line ending process in the color conversion process shown in FIG. 3;

FIG. 10 is a diagram illustrating a process for each pixel for discriminating a natural image / text image in the unprocessed row.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Luminance-density conversion part 3 Printer 101 Achromatic color judging means 102 Natural image / text image judging means 103, 106 Selector 104 Natural image processing interpolation operation means 105 Text image processing interpolation operation processing 107 Line buffer 108 Control means 201 Host device 401 processing window 402 target pixel 403 image area

Claims (13)

[Claims] 1. A color conversion device for performing color conversion using a look-up table, comprising: a determination unit configured to determine whether an image to be printed is a natural image or a non-natural image; Based on a data conversion process using a look-up table adapted to the natural image, or
A data conversion unit for performing any one of other data conversion processes using a look-up table adapted to an image other than the natural image. 2. The color conversion apparatus according to claim 1, wherein the image other than the natural image is a text image. 3. The color conversion according to claim 1, wherein the data conversion processing and the other data conversion processing each perform an interpolation operation on data read from the look-up table. apparatus. 4. The first determining means determines whether an image to be printed is an achromatic image. When the first determining means determines that the image is an achromatic image, the image is determined to be an achromatic image. And a second determining unit that determines whether the image is a natural image or an image other than a natural image. When the first determining unit determines that the image is not an achromatic image, the image is determined to be a natural image. The color conversion device according to claim 1, wherein 5. The image processing apparatus according to claim 1, wherein the second determining unit determines whether or not the image is a natural image based on a statistical feature of the image. Based on the difference from the average pixel value of the pixel of interest, it is determined whether or not the pixel of interest is in the character area, and among the pixels of the predetermined image area including the pixel of interest, the pixel has an achromatic pixel value. By comparing the pixel pattern with a predetermined pattern,
5. The color conversion device according to claim 4, further comprising: a second unit configured to determine whether the target pixel is in a grayscale area to determine whether the pixel is an image other than the natural image. 6. 6. A color conversion device for performing color conversion using a look-up table, comprising means for performing color conversion using a three-dimensional look-up table adapted for natural image processing, and adapted for text image processing. Means for performing color conversion using the one-dimensional look-up table described above. 7. A color conversion method for performing color conversion using a look-up table, comprising: determining whether an image to be printed is a natural image or an image other than the natural image; Performing one of a data conversion process using a look-up table suitable for an image and another data conversion process using a look-up table suitable for an image other than the natural image. Color conversion method. 8. The color conversion method according to claim 7, wherein the image other than the natural image is a text image. 9. The color conversion according to claim 7, wherein the data conversion processing and the other data conversion processing each perform an interpolation operation on data read from the look-up table. Method. 10. The determining step includes: a first determining step of determining whether an image to be printed is an achromatic image; and determining that the image is an achromatic image by the first determining step. And a second determining step of determining whether the image is a natural image or a non-natural image. In the first determining step, when it is determined that the image is not an achromatic image, the image is determined to be a natural image. The color conversion method according to any one of claims 7 to 9, wherein: 11. The second determining step includes: a first step of determining whether or not the image is a natural image based on a statistical feature of the image; Based on the difference from the average pixel value of the pixel of interest, it is determined whether or not the pixel of interest is in the character area, and among the pixels of the predetermined image area including the pixel of interest, the pixel has an achromatic pixel value. A second step of comparing the pattern of pixels with a predetermined pattern to determine whether or not the pixel of interest is in a gray scale area to determine whether or not the image is other than the natural image. The color conversion method according to claim 10, wherein: 12. A printing system for performing printing based on color-converted data, a color conversion apparatus for performing color conversion using a look-up table, wherein an image to be printed is a natural image or the natural image. Discriminating means for discriminating whether the image is other than an image, and, based on the discrimination by the discriminating means, data conversion processing using a look-up table adapted to a natural image
A data conversion means for performing any one of other data conversion processes using a look-up table adapted to an image other than the natural image; and a color conversion device comprising: A printing system, comprising: a printing means for performing printing by means of: 13. An image processing method for performing a process of determining a type of an image, comprising: determining a difference between a pixel value of a pixel of interest of an image to be output and an average pixel value of pixels around the pixel of interest. It is determined whether or not the pixel of interest is in a character area, and among pixels of the predetermined image area including the pixel of interest, a pattern of a pixel having an achromatic pixel value is compared with a predetermined pattern, and An image processing method, comprising: determining whether a target pixel is in a gray scale area; and determining whether the pixel is an image other than a natural image by including each step.