JPH07111602A - Image forming device - Google Patents

Abstract

(57) [Abstract] [Purpose] To eliminate the bleeding that occurs when filtering is performed for each color and to eliminate the occurrence of white spots in the erase processing other than black to improve image quality. [Structure] An image reading unit for reading image data, and RGB image data read by the image reading unit for YMCBk
The image processing unit includes an image processing unit for converting into image data and an image recording unit for executing image recording based on the YMCBk image data converted by the image processing unit.
An original document recognition unit 202 that recognizes the original document state for each color of the GB image data, and a different spatial filter process for each color based on the output data of the original document recognition unit 202, and a correction process is performed on the processed data. RGB filter 2 to execute
And 03.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image forming apparatus such as a digital color copying machine, and more particularly to recognizing an original image and performing different filter processing and correction processing for each color, or erasing processing by referring to peripheral data. The present invention relates to an image forming apparatus that executes image processing such as performing each color.

[0002]

2. Description of the Related Art As a conventional reference document related to the present invention, there is an "image processing device" disclosed in Japanese Patent Application Laid-Open No. 2-295358. This uses the minimum value of the luminance signal Y or the RGB signal for character / pattern separation recognition, determines the chromatic / achromatic color as the chromatic color when the RGB signal is not uniform, and based on these determination results, Each process such as is executed. That is, in the case of black character processing, edge enhancement, color residue removal, 400d
output in pi, and edge emphasis in case of color character output.
Output at 0 dpi, output smoothing at 200 dpi for halftone image processing, smoothing or through 200 for halftone image processing.
The output is controlled to be in dpi.

In the case of a black character, a character in a print image is expressed only by black ink, and in the case of a color character, solid data (not halftone dots) of at least one color and at least one character.
It is composed of color halftone dot data.

[0004]

However, in the conventional image processing apparatus as shown above,
There were the following problems. First, in general, halftone dot data is smoothed by a filter, while character information (solid data) is sharpened, but a different process for each color, for example, R is a smoothing process. When B is sharpened, bleeding occurs due to R data adhering to the periphery of B, which causes a problem of image deterioration.

Secondly, in the black character processing, the black character is printed in black, and the erase processing is performed on the portion other than black. However, when the ground color has a color, the erase portion is white. There is also a problem in that a dropout occurs and image deterioration is caused.

The present invention has been made in view of the above, and eliminates bleeding that occurs when filter processing is performed for each color and eliminates the occurrence of white spots in erase processing other than black. ， The purpose is to improve the image quality.

[0007]

In order to achieve the above object, the present invention provides an image reading unit for reading image data and an image processing for converting RGB image data read by the image reading unit into YMCBk image data. Means,
In an image forming apparatus having an image recording unit that executes image recording based on the YMCBk image data converted by the image processing unit, the image processing unit includes the R
Original recognition means for recognizing the original status for each color of GB image data, and different spatial filter processing for each color based on the output data of the original recognition means, and correction processing for the processed data. The present invention provides an image forming apparatus including a filter processing unit for

Further, the filter processing means executes smoothing processing on the halftone dot image.

Further, the filter processing means determines whether or not the correction processing can be executed based on the edge extraction amount.

Image reading means for reading the image data, and image processing means for converting the RGB image data read by the image reading means into YMCBk image data.
An image forming apparatus having image recording means for executing image recording based on the YMCBk image data converted by the image processing means and for printing black in one color,
The image processing means includes a detecting means for detecting the contour of a black character, an erasing means for independently erasing the peripheral data of the black character for each color, and a possibility of erasing by the erasing means for each color based on the peripheral data. The image forming apparatus is provided with a determining unit for determining.

Further, the judging means makes a judgment based on the edge extraction amount.

[0012]

In the image forming apparatus according to the present invention, the document recognition unit recognizes the document state for each color with respect to the RGB image data, and the filter processing unit differs for each color based on the output data of the document recognition unit. Spatial filter processing is executed, and correction processing is executed on the processed data to eliminate the occurrence of bleeding and emphasize and reproduce the contours of characters.

Further, the detecting means detects the contour of the black character, and the erasing means executes the erasing process of the peripheral data of the black character independently for each color based on the result of the judging means. Eliminate blank areas.

[0014]

Embodiments of the present invention will be described below with reference to the accompanying drawings. FIG. 1 is a block diagram showing the main configuration of a digital color copying machine according to the present invention. In the figure,
Reference numeral 101 denotes an image reading unit that reads image information as digital image data while scanning a document, 102 denotes an image processing unit that performs a predetermined processing process on the image data read by the image reading unit 101, and 103 denotes image processing. Part 1
An image recording unit that outputs the image data processed by 02 to recording paper.

FIG. 2 is a block diagram showing the detailed arrangement of the image processing unit 102. In the figure, 201 is an RGB γ correction unit that executes γ correction on the read data (RGB image signal) of the image reading unit 101 based on a preset look-up table, and 202 is each corresponding to the RGB image signal. A document recognition unit that changes the processing of blocks.

Reference numeral 203 denotes a N × N spatial filter corresponding to the image state of the image reading unit 101, and selects each processing such as smoothing processing, sharpening processing, or processing as it is. , RGB filter to execute, 20
Reference numeral 4 is a color correction (masking) unit for converting the RGB image signal of additive color mixture into YMC data of subtractive color mixture, and 205 is YMC to Bk for ensuring the gray balance in the intermediate density range.
Data of Y, M, C, B
This is a UCR unit that outputs one color of k in a frame sequential manner.
In addition, here, IMG is, among Y, M, C, and Bk,
Means one color.

Further, reference numeral 206 denotes a scaling processing unit for performing enlarging / reducing in the main scanning direction, or the same magnification processing, 207.
Is a processing unit that executes processing such as mirror processing, italic processing, shadowing processing, paint processing, movement processing, and repeat mosaic in the main scanning direction on the image signal as necessary. A CMYBk is executed by selecting a smoothing process, a sharpening process, or a process in the unprocessed state as it is, using an N × N spatial filter according to the frequency characteristic of the recording unit 103 and the image state. A filter, 209 is a CMYBk correction unit that performs γ correction in accordance with the γ characteristic of the image recording unit 103 and the image state, and 210 is a floor that performs quantization such as dither processing based on the gradation characteristic of the image recording unit 103. It is a tone processing unit.

In the digital color copying machine configured as described above, R (red), G (green) and B (blue) data of the original image are simultaneously input in parallel and Y (yellow). , M (magenta), C (cyan),
A color image is formed by converting the data into Bk (black) data and processing each color data in a frame sequential manner. In this case, since each signal of YMCBk is output in the frame sequential manner, the image reading unit 101 performs four scans during image formation. In addition, in the present embodiment, when the RGB data becomes large, it becomes white, and CMYB
The larger the k data, the darker it becomes.

FIG. 3 is a block diagram showing the configuration of the document recognition unit 202 shown in FIG. This document recognition unit 202
Is a color determination unit 301 which inputs RGB image data and outputs data for determining whether the input image data is halftone dot data, and a halftone dot data separated from the input image data and output. Halftone dot separating section 302, character determining section 303 for determining a character component for each color, and character determining section 303
And a character control unit 304 that outputs the character contour control data and the component for each color on the basis of the determination data.

Further, between the output terminal Q of the color judgment section 301 and the input terminals I1 to I3 of the halftone dot separation section 302, a line for executing a plurality of line processes so that the halftone dot judgment and the character judgment can be executed. The memories 305 and 306 are connected. Further, between the terminal SEL of the color determination unit 301 and the terminals SEL of the halftone dot separation unit 302 and the character determination unit 303,
Line memory 307 to match Q output and delay
Is provided.

Next, the operation of the document recognizing unit 202 having the above structure will be described. R of the color determination unit 301,
The RGB image data read by the image reading unit 101 is input to the G and B input terminals, respectively, and the data for determining whether the data is halftone dot data is output to the output terminal Q. In this case, as the determination signal, the minimum value of the image data, the luminance signal, or the G data is used. Further, from the SEL terminal, the RGB image data is referred to and color information is transmitted via the line memory 307 to the halftone dot separating unit 302.
And output to the character determination unit 303. In this case, the color information is RGB binary information (R, G, B, Y,
M, C, Bk), or further detailed information.

Further, the halftone dot separating section 302 executes a generally known halftone dot separating process. However, since the processing is executed by using the specific data, it may not be possible to determine that the image is a halftone image even though it is a halftone image. At this time, correction is performed based on the input SEL. That is, the correction is executed based on the variation state of the color information due to the input SEL.

As this method, referring to the change in the color of the pixels before and after, it is considered that it is not a halftone dot when continuously changing, and conversely, when it is discontinuous and a large number of dots are lined up. Judge that there is. Further, the output P is set to H when it is determined that one of the known halftone dot separation and the halftone dot determination detected by the correction is a halftone dot. Note that the discontinuity in the above means a case where the pixel is composed of a large number of 2 pixels, for example, 25 pixels or more out of 32 pixels.

Further, the character determination unit 303 performs known character determination processing, but Y for components other than the Bk component (QBk) is performed.
The component (QY), M component (QM), or C component (QC) is output. Note that in this case, for the character, only the outline of the character is output. The contour portion of the character may be performed independently for Y, M, C, and Bk, but it is preferable to develop the contour portion into Y, M, C, Bk, R, G, and B for processing.

The character control unit 304 also outputs the signals (QBk, QY, QM, Q) determined by the character determination unit 303.
Based on C), character contour control C3 and R component (Q
R), G component (QG), and B component (QB) are output. That is, when C3 is H, it is a character signal, and P
Is a halftone dot, H is a halftone dot, C3 is L, and
When P is L, the tone is continuous.

Table 1 below is a logical table in the above document recognition processing, and shows the character determination result, the character contour control C3, and the character control result for each color system of the document data.

[0027]

[Table 1]

FIG. 4 shows the RGB filter 20 shown in FIG.
3 is a circuit diagram showing the configuration of FIG. This RGB filter 203
As shown in the figure, it is composed of a filter R401, a filter G402, a filter B403, and a selector 404 provided for each color of R, G, and B. Each filter 4
01 to 403 have the same internal configuration but different connection configurations. IMGI is image data, C2 is a character erase signal, ED2 is an edge component of G data, and ED3 is B.
It is an edge component of data.

RGB filter 2 constructed as described above
The operation of 03 will be described. The filter R401 performs spatial filtering on the image data IMGI and outputs the image data IMGO and at the same time outputs the edge-extracted data ED. It also outputs a character erase signal C2. In addition, similar processing is executed in the other filters G402 and B403.

Further, the selector 404 sets C at the time of Y printing.
When R = H, CG = H, CB = H (black characters), output C
2 outputs the BD data as it is. When M = printing, CR = H, CG = H, CB = H, output C2
Outputs the RD data as it is. Also, except for the above conditions, L is output. By this operation, the erase control at the time of black character is executed. The erase control will be described in the γ correction unit.

Table 2 below is a truth table in the above processing. In Table 2, ED1 is an edge extraction component of R, and is set to "large" when the absolute value of edge extraction is larger than a certain threshold value, and conversely is set to "small" when it is small. Others are the same.

[0032]

[Table 2]

FIG. 5 is a circuit diagram showing the internal structure of the filter R401, the filter G402, and the filter B403. In the figure, 501 and 502 are line memories, 503 is a smoothing processing unit that executes smoothing processing, and 50 is a smoothing processing unit.
Reference numeral 4 denotes an edge extraction unit that executes an edge extraction process, 505 a delay circuit that executes a delay process, 506 an adder that adds outputs from the edge extraction unit 504 and the delay circuit 505, and 507 an overall unit of the filter. A control unit 508 for controlling, a smoothing processing unit 503, an adder 506,
A selector for selecting an output from the delay circuit 505, 5
Reference numeral 09 denotes a correction processing unit that executes correction processing on the output from the selector 508.

Next, the filter R constructed as described above
Operations of the filter 401, the filter G402, and the filter B403 will be described with reference to FIGS. FIG. 6 is an explanatory diagram showing the concepts of the smoothing process (b), the edge extracting process (c), and the sharpening process (d) for the read data (a). 7A and 7B show 3 × 5 matrix filters (coefficients) used in each process, where (a) is an edge extraction filter, (b) is a sharpening filter, and (c) is a smoothing filter.

The smoothing processing unit 503, in order to avoid the occurrence of moire on the halftone dot original, the filter shown in FIG. 7C is applied to the read data (a) as shown in FIG. 6B. To perform the smoothing process. In addition, the edge extraction in the edge extraction unit 504 is performed in order to correct the MTF (blur) of the optical lens system of the read data as shown in FIG.
For the read data (a) as shown in (c),
The edge component is extracted using the filter shown in (a).

Further, the delay circuit 505 adjusts the delay time because a pixel delay occurs when executing the smoothing process and the edge extraction. Next, the adder 506 adds the outputs of the smoothing processing unit 503 and the delay circuit 505. That is, the edge extraction output data and the delay output are added, and the sharpening processing result is output.

Further, the selector 508 inputs the data from the smoothing processing section 503 to the input terminal A, the sharpening data from the adder 506 to the input terminal B, and the delay circuit 505 to the input terminal C. Through data is input, and based on the signals of SEL0 and SEL from the control unit 507, any data of the input terminals A, B and C is selected and output to the output terminal Y. Further, the correction processing unit 509 receives the output Y of the selector 508 and determines whether to erase according to the output signal E of the control unit 507. It should be noted that here, erase means conversion into white data.

The control unit 507 also outputs SEL0, 1 based on the input data and selects the output Y of the selector 508. That is, as shown in the truth table of Table 3 below, the selector 5 is selected based on the states of the signal P and the signal CD1.
A selection signal is given to 08. Further, the control unit 507 outputs the signal E to the correction processing unit 509 based on the input data. When the signal E is H, the correction processing unit 509 erases the image data.

[0039]

[Table 3]

The truth table shown in Table 2 will be explained. ED2 and ED3 are edge extraction components of G and B, and the absolute value of edge extraction is compared with a predetermined threshold value to extract the edge. “Large” when the absolute value of is large
Is output, and conversely, "small" is output when the absolute value of edge extraction is small. Further, ED1 is an edge extraction component of R, and when this value is small in a predetermined threshold value (minus value), it outputs "large", and otherwise outputs "small".

Further, the filtering process will be described in detail. FIG. 8 is an explanatory diagram showing the filter processing according to the present invention. In FIG. 8A, (a) to (c) are read data shown by a cross-sectional view in the main scanning direction, (a) is read data R, (b) is read data G, and (c) is read data. B. Further, the read data R in (a) is character data output as a result of character determination, the read data G in (b) is halftone dot data, and the read data B in (c) is continuous tone data (background). . Therefore, in this case, in the signal in FIG. 4, the halftone dot data signal P is H and CR is H, and only the outline of character determination is determined to be a character. Other CG, CB
Is L.

Next, for each of the above read data,
(B) Edge extraction (d) to (f), (c) smoothing process (g) and (h), (d) sharpening process (i) and correction process (j) are executed. Hereinafter, these processes will be described in order. As shown in the figure, the edge component in the read data R of (a) is (d), and the read data G of (b) is (e) when only the edge component of (d) is shown. Further, the edge component of the read data B in (c) is almost 0 (very small) as in (f) because the amount of change is small.

Next, since the read data G in (b) is halftone dot data, smoothing processing is executed and (g) is output.
Further, since the halftone dot data signal P of the read data B in (c) is H, smoothing processing is executed and (h) is output.

Here, since the shaded portion of (d) and the edge component of (e) correspond to "large" shown in Table 2 above,
Erase the periphery as shown in (i) with respect to (g) and output. That is, the read data (a) becomes (i), the read data (b) becomes (j), and the read data (c) becomes (h).

That is, it is possible to eliminate the occurrence of blurring (blurring) when other color data is smoothed around the sharpened data in the character portion. Further, the ground color refers to the edge component and is not corrected. Although a simple linear filter is used in the above embodiment, a non-linear filter can also be applied.

FIG. 9 is a circuit diagram showing the configuration of the UCR unit 205 shown in FIG. 2. In the figure, Y, M, and C are output image data from the color correction unit 204. Also, CR,
CG and CB are outputs of the document recognizing unit 202, and are controlled by a delay circuit (not shown) so that the positional relationships of the image data match.

The operation of the UCR unit 205 in the above configuration is as follows. That is, (a) In the case of CR = L, CG = L, CB = L (image) Further, in order to adjust the gray balance, an appropriate blackening process is executed. (B) When CR = H, CG = H, and CB = H (black characters) Further, the UCR amount is set to 100%, and printing is performed in full black. (C) In cases other than the above (a) and (b) (colored characters) The UCR amount is turned off (0%) and output in three colors of Y, M, and C. That is, Bk data is not printed.

Therefore, as described above, the image data emphasizes color reproduction, outputs black characters in full black, and outputs color characters without inking processing.

FIG. 10 is a circuit diagram showing the configuration of the CMYBkγ correction section 209 shown in FIG.
01 is a γ table, 1002 is an inverter, 1003-
1 to 1003-n are AND circuits. It should be noted that C2 is an output signal of the filter unit, which is a color data erase processing signal for black characters.

CMYBkγ correction section 20 in the above configuration
The IMGI input to 9 is data for one color of CMYBk, and is the output of the CMYBk filter 208. When C2 is L, normal γ correction is executed,
When C2 is H, the output IMGO is all L. Here, when the output data are all H, the density is high (black in the case of Bk printing), and conversely, when all the output data is L, the density is low (white).

That is, the erasing process is executed by the γ-correction. In the case of a black character on a white background in the outline portion of the black character, since edge components of all color data exist, the erasing process is performed on the color data other than black. 、 M and C (G and R) for black letters on the color ground (eg, yellow ground)
Since it is the edge component of the color data of, the color data of the contour portions M and C of the black character is erased.

An example of the above processing is shown in FIGS.
As shown in the figure, the correction processing is executed even when different processing is performed for each color by the filter, so that the contour of the character (edge emphasis) can be reproduced without the color bleeding. In addition, because the peripheral data is referenced and the erase process is controlled for each color, even for black characters on the color ground,
White spots do not occur around black characters.

[0053]

As described above, according to the image forming apparatus of the present invention, the document state is recognized for each color, the different spatial filter processing is executed for each color based on the recognition data, and the processed data is processed. To perform the correction process for
The occurrence of bleeding can be eliminated, the outline of a character can be reproduced clearly, and the outline of a black character is detected, and the peripheral data of the black character is erased independently for each color. Can be eliminated and the image quality can be improved.

[Brief description of drawings]

FIG. 1 is a block diagram showing the main configuration of a digital color copying machine according to the present invention.

FIG. 2 is a block diagram showing a detailed configuration of an image processing unit shown in FIG.

FIG. 3 is a block diagram showing a configuration of a document recognition unit shown in FIG.

4 is a circuit diagram showing a configuration of an RGB filter shown in FIG.

5 is a circuit diagram showing an internal configuration of a filter R, a filter G, and a filter B shown in FIG.

FIG. 6 is an explanatory diagram showing the concept of smoothing processing (b), edge extraction (c), and sharpening processing (d) for read data (a) according to the present invention.

FIG. 7 is a 3 × 5 matrix filter (coefficient) used in each process according to the present invention.

FIG. 8 is an explanatory diagram showing a filtering process according to the present invention.

9 is a circuit diagram showing a configuration of a UCR unit shown in FIG.

10 is a circuit diagram showing a configuration of a CMYBkγ correction section shown in FIG.

FIG. 11 is an explanatory diagram showing a first processing example according to the present invention.

FIG. 12 is an explanatory diagram showing a second processing example according to the present invention.

[Explanation of symbols]

101 image reading unit 102 image processing unit 103 image recording unit 202 document recognition unit 203 RGB filter 205 UCR unit 209 CMYBkγ correction

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location G06T 7/00 H04N 1/409 1/46 8420-5L G06F 15/66 310 9191-5L 15/68 400 A 7459-5L 15/70 310 4226-5C H04N 1/40 101 C 4226-5C 101 D 4226-5C 1/46 Z

Claims (5)

[Claims] 1. An image reading means for reading image data,
An image forming apparatus having image processing means for converting RGB image data read by the image reading means into YMCBk image data, and image recording means for performing image recording based on the YMCBk image data converted by the image processing means. In the image processing means,
A document recognition unit for recognizing a document state for each color of image data, and a different spatial filter process for each color based on the output data of the document recognition unit, and a correction process for the processed data. An image forming apparatus comprising: a filter processing unit. 2. The image forming apparatus according to claim 1, wherein the filter processing unit executes a smoothing process on the halftone image. 3. The image forming apparatus according to claim 1, wherein the filter processing unit determines whether or not the correction processing can be performed based on the edge extraction amount. 4. Image reading means for reading image data,
Image processing means for converting the RGB image data read by the image reading means into YMCBk image data, and image recording is executed based on the YMCBk image data converted by the image processing means, and black is printed in one color. In the image forming apparatus having the image recording means, the image processing means detects the contour of a black character,
An image forming apparatus comprising: an eraser for independently performing erase processing on peripheral data of black characters for each color; and a determiner for determining whether or not the erase processing by the erase means is possible for each color based on the peripheral data. apparatus. 5. The image forming apparatus according to claim 4, wherein the determination unit makes a determination based on an edge extraction amount.