JP2000032286A - Image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enhance the processing performance of an image in matching with a property of the image with a simple configuration by providing a color space conversion means that converts an RGB color signal into a YIQ color signal to the processor to conduct compression processing. SOLUTION: An identification device 1003 receives input image data consisting of RGB signals 1007 to use a difference between a maximum value and a minimum value in a 3×3 matrix for a difference threshold level and to check each pixel value in the matrix. The identification device 1003 identifies whether or not an area is a character area based on a property of characters where each has a comparatively high pixel value and a large pixel value difference from surrounding pixel values to produce an identification signal 1008. An image correction device 1004 converts the RGB signal 1007 into a YIQ signal and applies emphasis processing to character pixels in terms of the YIQ signal. In this case, four signals consisting of the RGB signals 1007 and the identification signal 1008 are converted into the YIQ signal by using prescribed equation. Since parameters correspond to a luminance color difference in a YIQ space, the image quality including strength of strokes of characters and colors is easily adjusted in a sense close to human sense.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus used for a facsimile, a printer, and the like.

[0002]

2. Description of the Related Art Conventionally, in a device for outputting an image, such as a facsimile or a printer, an image compression / expansion technique has been used to compress an image and effectively use a transmission line and a memory.

As techniques for responding to these demands, there are, for example, Japanese Patent No. 2555338 (referred to as Document 1) and Japanese Patent Publication No. 7-97829 (referred to as Document 2).

[0004] The method disclosed in Reference 1 converts a color signal, that is, three primary color signals (RGB signals) into a luminance signal (Y signal).
And a chrominance signal (IQ signal) is generated, and the block size is kept in accordance with the amount of change of the chrominance signal in the processing block, or the block is divided into two, and an averaging process is performed.

[0005] The method disclosed in the above reference 2 directly outputs a luminance / color difference signal (YIQ) from three primary color signals (RGB signals).
Rather than obtaining the signals, the three primary color signals are converted into density signals (CMY signals), and then converted into luminance / color difference signals, thereby compressing the luminance signal and the color difference signal. In this compression, the compression ratio for the color difference signal is set higher than the compression ratio for the luminance signal.

[0006]

However, in both References 1 and 2, the color space is converted from three primary color signals and density signals to a luminance / color difference space. When the output is increased or the saturation is increased, a separate conversion process needs to be performed.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned conventional problems, and it is possible to convert a color space in accordance with the characteristics of an image with a simple configuration, thereby improving the processability of the image. It is an object of the present invention to provide an image processing apparatus which can improve the image quality and reduce the image quality deterioration at the time of compression processing of an image in a conversion color space.

[0008]

The invention according to claim 1 is
Color space conversion means is provided for converting the first color signal into a second color signal using a first color signal and a signal representing information different from the first color signal.

According to the first aspect of the present invention, nonuniform color signal conversion can be performed, so that image processing is improved.

[0010]

[1] Hereinafter, a first embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing the overall configuration of a digital color copying machine to which the present invention is applied.

This digital color copying machine reads an original image to be copied as color image data, and outputs three primary color signals representing the color image, so-called RGB signals 1007.
, A line memory 1002 for storing RGB signals (input image data) 1007 output from the scanner 1001 in units of lines, a character area in a reference area pixel in the line memory 1002, and a result of the identification. 1003 that outputs an identification signal 1008 representing
An image correction device 1004 that corrects the GB signal with an identification signal 1008 from the identification device 1003 to generate a corrected image signal 1009, and converts a corrected image signal (RGB signal) 1009 generated by the image correction device 1004 into a color space signal (density). A color conversion device 1005 for converting the image into a signal (a so-called CMY signal) 1010 and a printer engine 1006 for forming an image in accordance with the color space signal 1010 obtained by the color conversion device 1005.

In this embodiment, if RGB = 0, black, R
White if GB = 255, white if CMY = 0, CMY = 25
5 is described as black.

Next, the operation of the identification device 1003 will be described with reference to the drawings.

The identification device 1003 operates to identify a character area. Here, the character is a region surrounded by a region having a relatively large pixel value and a large pixel value difference from the surroundings. For example, if the pixel value difference, which is the density difference between adjacent pixels, is large and the pixel value is small, a character is displayed. The pixel value is small,
Non-character if the pixel value difference is small. And so on.

As a configuration, for example, as shown in FIG.
By comparing the difference between the maximum value and the minimum value in the × 3 matrix with a difference threshold value, it is determined whether the region satisfies the character pixel level by comparing the pixel value with the luminance threshold value. , And “0” for non-character areas.

Referring to the operation of FIG. 2, a line memory 1002 and a selector 1003-2 are used to retrieve data in a 3 × 3 matrix.

In the operation of the line memory, the data can be read and transferred without stopping the scanner 1001 by sequentially storing the data from the scanner in the line memory from which the data has not been extracted.

In the discriminating apparatus 1003, since the input signal is RGB, the three signals are converted to 1003-1-1, 103-1.
A luminance signal is generated using the averaging circuit of 03-1-2, 1003-1-3, and 1003-1-4.

The selector 1003-2 sets the luminance signal of the upper pixel of the matrix to 1003-3, the luminance signal of the lower pixel to 1003-4, and the luminance signal of the middle pixel to 1003-3.
5, the RGB signals of the middle pixel are represented by 1007-1 and 1007, respectively.
-2, 1007-3. The luminance signal of each line is output through a flip-flop D-FF to 3
It is taken out pixel by pixel.

In this configuration, the luminance signal of the 3 × 3 matrix is
Maximum value comparator 1003-6, minimum value comparator 1003-7
And the maximum and minimum values in the matrix are
(Maximum value−minimum value) is obtained as a difference. The calculated difference value is compared with the difference threshold value 1003-9 by the comparator 1003-10.
If it is larger than −9, “1” is output, otherwise “0” is output.

On the other hand, the luminance signal of the central pixel of the matrix is "1" if the luminance signal is smaller than the luminance threshold 1003-11 by using the luminance threshold 1003-11 and the comparator 1003-12, and is "0" otherwise. Is output. Memory 1
003-13 stores and outputs the output of the comparator 1003-11 when the comparator 1003-10 is "1", and outputs the stored content as the identification signal 1008 when the comparator 1003-10 is "0".

The RGB signals 1007 are shown in FIG. 2 as 1007-1, 1007-2, and 1007 for explanation.
-3.

As shown in FIGS. 3 and 4, the comparator 10
03-10, output of 1003-11 and memory 1003-
13 outputs "1" as an identification signal only in a character area.

Next, the pixel correction device 1004 will be described with reference to FIG.

The pixel correction device converts the RGB signals into luminance / color difference signals, and performs character pixel emphasis processing on the luminance / color difference signals.

In this example, the following equation (Equation 1) is used as the luminance / color difference conversion, but the conversion to a luminance / color difference system is not limited to this. In the pixel correction device 1004,
Since character pixel enhancement processing is performed using the identification signal, four signals are converted into three signals as shown in the following equation (Equation 2). The inverse transformation is shown in the following equation (Equation 3).

[0028]

(Equation 1)

[0029]

(Equation 2)

[0030]

(Equation 3)

For example, offset (parameter) α,
Regarding β and γ, if α = −10 and β = γ = 0, when the identification signal is “1”, the brightness of the character is weakened, and the character is black and has a contrast. If β = −10 and α = γ = 0, the G signal of the character is weakened and the R and B signals are strengthened, so that the color can be changed without changing the luminance of the character.

FIG. 5 shows a circuit configuration example. RGB signal 10
07-1, 1007-2, and 1007-3 are converted to YIQ signals using a shift operation unit (shift conversion) and an adder (ADC). At this time, the offset α · β · γ1004
-1, 1004-2, and 1004-3 are added. The addition is performed only when the image signal is a character, that is, when the identification signal 1008 is “1”.

After being converted again into RGB signals, the out-of-range correction of the signals is performed in steps 1004-4, 1004-5, and 1004-6.
And the corrected image signals 1009-1, 1009-2, 1
Output as 009-3. Note that the out-of-range correction is performed by adding the offset so that the RGB signals have levels “0” to
If the level is outside the range of “5”, the level is corrected to “0” if the level is lower than “0”, and to 255 if the level is higher than “255”.

FIGS. 6 and 7 show conversion examples. FIG.
By operating only, the pixel value of the entire character is lowered (corresponding to black), and the character is emphasized black. FIG. 7 changes the color from gray to green without changing the brightness of the entire character by operating only β.

Next, the color converter 1005 converts the RGB signals into CMY signals for output by the printer engine 1006. The conversion formula is C = 255-R, M = 255.
-G, Y = 255-B.

As described above, by performing the conversion into the YIQ signal using the RGB signal and the character identification signal different from the RGB signal, the image quality of the character is adjusted in the YIQ space, that is, the character is adjusted. The strength and color can be adjusted. That is, character emphasis processing or the like can be performed on the converted color signal, and image processing can be easily performed on a signal close to human sense such as luminance / color difference.

In the above embodiment, the identification signal is set to "1".
Although the binary signal of “0” is used, the degree of identification (for example,
A multi-value signal indicating “5: character” →→ “−5: photograph”) may be generated and used.

Further, in the above embodiment, the conversion is performed to the YIQ space of the shift operation and the matrix operation by the adder. However, other color spaces such as the L * a * b * space which is more complicated to convert may be used. Instead of a simple addition / subtraction of the offset, the offset value may be adjusted according to a value in a color space, a surrounding pixel value, or the like, or a user or the like may arbitrarily specify an adjustment amount. .

Further, in the above embodiment, the color signals (RGB
Although the signal / character identification information is used as a signal different from the color signal, the signal different from the color signal is not limited to this, and a signal for performing desired image processing may be used.

[2] The second embodiment will be described.

The second embodiment is a modification of the first embodiment. A color judgment device 1011 is provided as shown in FIG. 8, and an achromatic judgment signal 1012 output from the achromatic judgment device 1011 is used as an image. A configuration in which image correction is provided to the correction device 1013 is conceivable.

An achromatic determination device 101 for the first embodiment
1, achromatic determination signal 1012, image correction device 1013,
Since the corrected image signal 1014 is different, it will be described below.

The achromatic determination device 1011 includes a comparator 101
At 1-1 and 1011-2, the R and G signals and the G and B signals are compared, and if they match, "1" is output. That is, R =
The achromatic signal 1012 outputs “1” when G = B, and outputs “0” otherwise. An achromatic color is a color other than gray, white, and black.

The image correction device 1013 performs the conversion represented by the following equation (Equation 4) and the inverse transformation represented by the following equation (Equation 5).

[0045]

(Equation 4)

[0046]

(Equation 5)

For example, α0 = α1 = −10, β0 = γ0
When = β1 = γ1 = 0, the achromatic image and the character image are darkened as a whole, and the achromatic character image is further darkened, so that the chromatic image is relatively bright and the contrast with the character can be obtained. .

FIG. 9 shows a configuration example of the image correction device 1013. Compared to the configuration of FIG. 5, the offset α0 for the identification signal,
101-1, 1013-2, 101 which are β0 and γ0
3-3, 1013-4, 1013-5, and 1013-6, which are offsets α1, β1, and γ1 for achromatic signals, are converted to YI.
The difference is that it is added to the adder during Q conversion. Converted YI
Inverted RGB corrected image signals 1014-1 and 101 from Q
4-2 and 1014-3 are generated and output.

FIG. 10 shows a conversion example of the second embodiment.
That is, by adding the achromatic signal 1012, the width of the conversion can be increased.

The identification signal 1008 and the achromatic signal 10
If 12 is not independent and converted using the signal of AND of both, it is possible to change the color of only chromatic characters without changing achromatic characters, or to add color only to achromatic characters. it can.

[3] A third embodiment will be described.

As shown in FIG. 11, a CMY / YIQ converter 2008 that corrects an image with the identification signal 2007 and the CMY signal 2004 and converts it into a YIQ signal, and an image compressor 201 that compresses the converted YIQ signal 2009
0, page memory 2012 for storing compressed data 2011, data 2011 compressed from page memory
And a YIQ / CMY converter 2015 for converting the decoded image signal 2014 into a CMY signal.
16 is supplied to the printer engine 2017.

The other structure is the same as that of the first embodiment.

That is, the CMY signal 2004 is held in the line memory 2005, the identification signal 2007 is generated by using the identification device 2006, and the CMY / YIQ conversion device 2008 generates Y by using the CMY signal 2004 and the identification signal 2007.
An IQ signal 2009 is generated, and compressed data 2011 is generated by the image compression device 2010 and stored in the page memory 2012.

Subsequently, the image decoding apparatus 2013 reads out the compressed data 2011 read from the page memory 2012 to generate an image decoding signal 2014, and converts it into a decoding CMY signal 2016 by the YIQ / CMY conversion apparatus 2015. Output and obtain copy manuscript.

Since the copying machine performs one to a plurality of copying operations by reading out the compressed data stored in the page memory 2012, the copying operation can be performed by a single scanning operation even for a plurality of copying operations.

In this embodiment, if RGB = 0, black, R
White if GB = 255, white if CMY = 0, CMY = 25
5 is described as black.

Each processing block will be described with reference to the drawings. The scanner 2001 and the color conversion device 2003 are the same as those in the first embodiment, and the line memory 2005 is the same except that the CMY signals are stored instead of the RGB signals. Description is omitted because there is.

FIG. 12 shows an identification device 2006, which is basically the same as that of the first embodiment, except that the CMY (density) signal is used instead of the RGB (luminance) signal.
6-12 converts the density signal 2006-5 to a density threshold 2006
The difference is that "1" is output if the value is larger than 11, and "0" is output otherwise. The identification is the same as the 1st order, but
If the pixel value is large and the pixel value difference is large, it is a character, and if the pixel value is large and the pixel value difference is small, it is a non-character.

FIG. 13 shows a CMY / YIQ converter. The conversion from CMY to YIQ is realized by replacing the input of the above equation (2) with RGB to CMY, and implementing the following equation (6). .

[0061]

(Equation 6)

Using the shift operation and the adder, the CMY signal and the offset α, β, γ2008-1, 2008-2,
The conversion from 2008-3 to the YIQ signal is the same as that of the YIQ conversion unit of the first embodiment. But the comparator 200
8-4 and out-of-range correction 2008-5, 2008-6, 20
The difference is that the YIQ signal 2009 is corrected using 08-7.

The comparator 2008-4 outputs "1" when the CMY signals are equal, and outputs "0" otherwise. The out-of-range correction 2008-5 corrects a case where the Y signal is out of the range due to the addition of the offset α2008-1. The out-of-range corrections 2008-6 and 2008-7 are offset β200
Correction when the IQ signal is out of the range due to the addition of 8-2 and γ2008-3, and correction for making the color difference component IQ signal “0” when the output of the comparator is “1”, ie, achromatic color. Tables showing the operation of the out-of-range correction are shown in FIGS.

With this configuration, the character image area is emphasized by offsetting. However, in the case of achromatic characters, the color difference deviation due to the emphasis is corrected and returned to achromatic, so that achromatic characters have only density and chromatic characters have density. And the saturation can be adjusted.

Next, the image compression apparatus 2010 will be described with reference to FIG.

Generally, it is known from the human visual characteristics that the information of the color difference signal is less noticeable even if quantized spatially coarsely, compared to the luminance and density signals.
09-2 and 2009-3 are subjected to 4-pixel averaging processing in averaging sections 2010-1 and 2010-2 via flip-flops D-FF, and the compression section 2010-3 spatially separates the IQ signal by one pixel for every four pixels. The image signal is compressed by extracting it roughly. The decoding of the compressed data 2011 is performed by extracting the Y signal in pixel units by the image decoding device 2013 and outputting the same value of the IQ signal as four pixels.

The conversion of the decoded YIQ signal 2014 into CMY is performed by a YIQ / CMY converter 2015 shown in FIG. 18, which is obtained by substituting the output of the above equation (3) from RGB to CMY. 7) The CM that performs the out-of-range corrections 2015-1, 2015-2, and 2015-3
Correction is made when Y falls outside the range of levels 0 to 255. That is, if it is smaller than “0”, “0”, “25”
If it is larger than "5", it is clipped to "255".

[0068]

(Equation 7)

From the CMY signal 2004 to the compressed data 201
FIG. 19 shows a generation process and a decoding process of No. 1 after the CMY signal of 96 bits (4 pixels) is converted into a YIQ signal, the Y signal is not changed, the IQ signal is averaged, and the data of 4 pixels is converted into one data. It is compressed to 50 bits by representing it with data.

In the decoding process, the Y signal is extracted for four pixels, and the IQ signal is output as a representative value of the same data for four pixels. Then, the decoded YIQ signal 2014 is converted into a CMY signal 2016 in pixel units.

The compressed and decoded signal is output by the printer engine 2017. Printer engine 20
When an electrophotographic printer is used in No. 17, the output characteristic of the engine is generally unstable in the low density region as shown in FIG.

For this reason, as shown in FIGS. 21 and 22, in the case of a photographic image, processing for preserving the density spatially such as error diffusion is performed, and the density is visually stored by outputting using a value that can be stably printed. However, if error diffusion or the like is used, the character image loses its shape, and the character image is noticeably deteriorated.

Therefore, when the IQ signal is spatially coarsely sampled and compressed, the characters can be recorded with little deterioration although the density is slightly different by emphasizing the IQ signal at the time of YIQ signal conversion. For example, assuming that a density value of 60 or more can be stably recorded, there is no character emphasis (α = β =
In the case of γ = 0), the data that could be stably recorded before compression becomes data that could not be stably recorded due to the averaging of the IQ components due to compression.
(β = 60, γ = −60) Data that can be stably recorded even after decoding.

Even if all the density values can be stably recorded, a change in the tint can be suppressed by averaging the IQ components. More specifically, in FIG. 19, when the character is not emphasized, an M component that is not in the original image is generated and the color tone is reduced. However, when the character is emphasized, blurring is slightly generated in the surroundings, but recording can be performed without changing the color.

As described above, according to the third embodiment, C
By converting to a YIQ signal using the MY signal and the identification signal, not only can the color and density of characters be adjusted,
When the IQ (color difference) component is roughly sampled and compressed in comparison with the Y (density) component, the deterioration of the character shape and color due to the compression can be suppressed.

Further, since the deterioration of the character image quality can be suppressed, it is possible to compress the image with the same image quality at a higher compression ratio as compared with the case where this processing is not used.

[4] A fourth embodiment will be described.

This is a modification of the third embodiment.
The configuration shown in FIG. The basic configuration is the same as that of the third embodiment, except that a CMY / YIQ conversion device 20081,
The difference lies in the configuration of the YIQ / CMY conversion device 20151 and that the compressed data 20111 handled by the image compression device 20101 and the image decoding device 20131 also compresses the identification signal 2007.

First, the CMY / YIQ converter 20081
Performs the normal YIQ conversion shown in the following equation (Equation 8).

[0080]

(Equation 8)

The image compression device 20101 and the image decoding device 20131 are basically the same as those in the third embodiment.
As shown in FIG. 24, the identification signal 2 is added to the compressed YIQ signal.
The difference is that compression / decoding processing is performed as 36-bit compressed data 201111 by adding a bit. That is, YIQ
FIG. 24 shows how the signal is compressed and decoded.

The YIQ / CMY converter 20151 calculates the CM from the YIQ signal and the identification signal as shown in the following equation (Equation 9).
Convert to Y signal.

[0083]

(Equation 9)

An example of the structure is shown in FIG.
0151-1, β20151-2, and γ20151-3 are ANDed with the identification signal 2007 and the state signal 20151-6 of the density signal 20141-1 (Y), and the YIQ signal 2
After that, the above equation (7) is executed.

The state signal 20151-6 is supplied to the comparator 201
At 51-4, the density signal 104414-1 (Y) is compared with the background density threshold 20151-5, and
If it is smaller than 5, it outputs "0", and if it is larger, it outputs "1".

According to this configuration, no offset is added to the background area having a value equal to or less than a certain value, and blurring around the character can be suppressed. FIG. 26 shows an operation example when the background density threshold is set to “10”.

Note that out-of-range corrections 20151-7, 201
The operations of 51-8 and 20151-9 are the same as those of the third embodiment shown in FIG.
Similarly, the CMY signal is clipped at 8 bits.

As described above, by performing character enhancement processing on data compressed / decoded on the YIQ color signal at the time of YIQ / CMY conversion, deterioration in image quality due to compression / decoding processing on the YIQ signal is reduced. can do.

In this embodiment, the YIQ / CMY conversion is performed using the compressed identification signal.
A configuration in which an identification signal is generated on a signal and used for YIQ / CMY conversion may be adopted. Furthermore, YIQ / C after decryption
The image was emphasized only at the time of MY conversion.
It is also possible to adopt a configuration in which correction processing such as character emphasis processing is performed both at the time of MY / YIQ conversion and at the time of YIQ / CMY conversion after decoding.

[0090]

As described above, according to the present invention,
The color space conversion means for converting the first signal into another color signal using the first color signal and a signal different from the first color signal is provided, so that the configuration can be adapted to the characteristics of the image with a simple configuration. It is possible to provide an image processing apparatus capable of performing color space conversion, thereby improving the processability of an image, and reducing image quality degradation during image compression processing in a converted color space.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a first embodiment.

FIG. 2 is a block diagram illustrating a configuration of an identification device according to the first embodiment.

FIG. 3 is a diagram for explaining the operation of the identification device in the first embodiment.

FIG. 4 is a diagram for explaining the operation of the identification device in the first embodiment.

FIG. 5 is a block diagram illustrating a configuration of an image correction device according to the first embodiment.

FIG. 6 is a diagram for explaining the operation of the image correction device of FIG. 5;

FIG. 7 is a diagram for explaining the operation of the image correction device of FIG. 5;

FIG. 8 is a block diagram showing a configuration of a second embodiment.

FIG. 9 is a block diagram illustrating a configuration of an image correction device according to a second embodiment.

FIG. 10 is a diagram illustrating the operation of an achromatic color determination device and an image correction device according to a second embodiment.

FIG. 11 is a block diagram showing a configuration of a third embodiment.

FIG. 12 is a block diagram illustrating a configuration of an identification device according to a third embodiment.

FIG. 13 is a block diagram illustrating a configuration of a CMY / YIQ conversion device according to a third embodiment.

FIG. 14 is a diagram for explaining the operation of the out-of-range correction unit in FIG. 13;

FIG. 15 is a diagram for explaining the operation of the out-of-range correction unit in FIG. 13;

FIG. 16 is a diagram for explaining the operation of the out-of-range correction unit in FIG. 13;

FIG. 17 is a block diagram illustrating a configuration of an image compression device according to a third embodiment.

FIG. 18 is a block diagram illustrating a configuration of a YIQ / CMY conversion device according to a third embodiment.

FIG. 19 is a diagram for explaining the operation from the CMY / YIQ converter to the YIQ / CMY converter in the third embodiment.

FIG. 20 is an explanatory diagram of output characteristics of a printer engine according to the third embodiment.

FIG. 21 is a view for explaining the operation of the printer engine in the third embodiment.

FIG. 22 is a view for explaining the operation of the printer engine in the third embodiment.

FIG. 23 is a block diagram showing a configuration of a fourth embodiment.

FIG. 24 is a diagram for explaining operations of the image compression device and the image decoding device in the fourth embodiment.

FIG. 25 is a block diagram illustrating a configuration of a YIQ / CMY conversion device according to a fourth embodiment.

FIG. 26 is a view for explaining the operation of the YIQ / CMY conversion device of FIG. 25;

[Explanation of symbols]

 1001 Scanner 1002 Line memory 1003 Identification device 1004 Image correction device 1005 Color converter 1006 Printer engine 1007 RGB signal 1008 Identification signal 1009 Corrected image signal 1010 Color space signal

 ──────────────────────────────────────────────────続 き Continued on the front page F-term (reference) 2C262 AA24 AA26 AA27 BA01 BA20 DA17 EA07 5C057 AA06 AA07 CA01 CE00 DA04 DA16 DC01 EA00 EA01 EA02 EA07 EA12 EL01 FB08 FE06 GF08 GH03 GJ01 GJ02 GJ09 PG03 MP025 TT06 5C079 HB01 HB02 HB12 LA06 NA02 NA27

Claims (1)

[Claims] 1. A color space conversion means for converting a first color signal into a second color signal using a first color signal and a signal representing information different from the first color signal. An image processing apparatus characterized by the above-mentioned.