JP2002281313A - Image processor and image forming apparatus having the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processor to make the edge of an image drawn in a dot area possible to be viewed more clearly. SOLUTION: When an object is judged as the one in the dot area and as the one adjacent to an edge pixel to constitute the edge of the image such as a character image, smoothing is performed to data of the object pixel by using the respective smoothing filters 481 to 488, the data to which the smoothing is performed by the filter by which the smoothing is performed without reference to the data of the edge pixel is outputted as data after the smoothing of the object pixel among pieces of data to which the smoothing is performed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image processing apparatus, and more particularly to a technique for suppressing deterioration of image quality in an image forming apparatus for forming an image based on digital image data.

[0002]

2. Description of the Related Art In an image forming apparatus for forming an image based on digital image data, various types of image processing are performed on pixels of a corresponding portion according to the type of the image, for example, a character image or a halftone image. Thus, the image quality has been improved. More specifically, a smoothing process (smoothing process) is performed on a pixel determined as a halftone dot region to prevent occurrence of moire, and an edge enhancement process is performed on a pixel determined as an edge region. It is common.

[0003] As a method of the smoothing processing, a filter composed of, for example, 3 pixels * 3 pixels centering on a pixel to be subjected to the smoothing processing (hereinafter referred to as "target pixel") is set and included in the filter. 2. Description of the Related Art A method of creating data of a smooth reproduced image by performing a moving average by weighted addition on the density value of each pixel has been conventionally known.

[0004]

However, when a conventional image processing apparatus is used, there is a problem that a character image drawn in a halftone dot region may not have sharp edges. . That is, conventionally, for all pixels determined to be a halftone dot area, smoothing processing is performed using the same filter with each pixel as a target pixel. Therefore, when a filter of 3 pixels * 3 pixels is used, for example, FIG. Assuming that pixel 901 is the target pixel in the image shown in FIG.
One pixel (902, 903, 904) will be included. On the other hand, if the target pixel is the pixel 905, the surrounding pixels do not include any pixels in the character image area, that is, only the pixels in the halftone dot area.

In general, a character image is composed of pixels having extremely high density continuously. In a halftone dot region, pixels having high density (pixels shaded in FIG. 15) are replaced with pixels having low density (shaded lines). (Pixels not subtracted with) in many cases. Therefore, when the density value is calculated using the same filter, the density value of the pixel 901 whose peripheral pixels include the pixels of the character image area is calculated as the pixel 90 whose peripheral pixels are only the halftone area.
It will be higher than 2. That is, pixel 9 in FIG.
Looking at 03, 901, and 905, the density gradually decreases in that order, and as a result, the density gradient from the edge part of the character image to the halftone dot area becomes loose,
Edges may not look sharp, or in other words, may appear blurry.

The above problem is not limited to an image drawn in a halftone dot area. For example, when an image such as a character is drawn in a flat density area such as a photographic image, the density flat area is more likely to be displayed. This may also occur when performing a smoothing process on the area in order to reproduce smoothly. The present invention has been made in view of the above-described problems, and has been made in consideration of the above-described problems, and has been made in consideration of an image processing capable of suppressing deterioration of image quality by performing processing so that edges of a character image or the like drawn in a halftone dot area or the like look sharper. It is an object to provide an apparatus and an image forming apparatus using the image processing apparatus.

[0007]

In order to achieve the above object, an image processing apparatus according to the present invention weights data of pixels within a predetermined range including a target pixel, and outputs data of the weighted pixels. An image processing apparatus comprising a smoothing means for smoothing the data of the target pixel based on (i), wherein each pixel within a predetermined range including the target pixel is a pixel existing in a character image area or not. Character image area determining means for determining, the smoothing means, when the target pixel is determined not to be in the character image area, and when there is a pixel determined to be in the character image area within the predetermined range, It is characterized in that the weighting amount for the pixel is made smaller than the weighting amount when the pixel is not in the character image area.

[0008] The image processing apparatus further includes a halftone dot region determining means for determining whether or not the pixel exists in the halftone dot region. It is characterized in that the weighting amount for a pixel determined to be in the character image area is smaller than the weighting amount when the pixel is in the halftone dot area. Further, a weighting amount for data of a pixel determined to be within the character image area is zero.

Here, "the weighting amount is zero" means that the data of the pixel having the weighting amount of zero is not referred to in the processing when the data is weighted. Further, within the predetermined range, a plurality of types of filters having a weighting amount for pixels having a predetermined positional relationship with respect to the target pixel are provided with a plurality of types of filters, and the smoothing unit includes, among the plurality of types of filters, It is characterized in that smoothing is performed using a filter configured so that the weighting amount for pixels determined to be within the character image area becomes zero.

According to the present invention, there is provided an image forming apparatus for forming an image based on data smoothed by the image processing apparatus, wherein the image processing apparatus includes the image processing apparatus according to the present invention. Features.

[0011]

Embodiments of an image processing apparatus and an image forming apparatus according to the present invention will be described below with reference to the drawings. (1) Overall Configuration of Image Forming Apparatus FIG. 1 is a schematic cross-sectional view showing the overall configuration of a full-color copying machine (hereinafter, simply referred to as “copier”) 1 as an example of an image forming apparatus.

The copying machine 1 uses an image forming unit 3 using digital image data obtained by reading an original by an image reading unit 200.
00 to form an image, the image reading unit 200
An automatic document feeder 100 is provided at the upper part of FIG.
Normally, the document conveyed to the image reading position by the automatic document feeder 100 is read by the image reading unit 200, and the obtained image data is transferred to the image forming unit 300, and the image forming unit 300 prints an image on a recording sheet. Form. Of course, the external interface 207 allows connection with an external device such as a personal computer (PC). Thereby, a scanner function of outputting image data obtained by reading by the image reading unit 200 to an external device,
It is possible to realize a printer function of forming an image in the image forming unit 300 using image data input from an external device.

The automatic document feeder 100 includes a document tray 1
The document set at 01 is transported to the image reading position of the image reading unit 200, and after reading the document, the document is discharged onto the document discharge tray 103. The document transport operation is performed according to an instruction from an operation panel (not shown).
The document discharging operation is performed according to a reading completion signal from the image reading unit 200. When a plurality of originals are set, these control signals are continuously generated, and the operations of conveying, reading, and discharging the originals are sequentially performed.

In the image reading section 200, the original glass 208
The original placed above is irradiated with an exposure lamp 201, and reflected light is imaged on a CCD sensor 204 via a mirror group 202 including three mirrors 2021 to 2023 and a lens 203. Exposure lamp 201 and first mirror 202
1 is driven by a scan motor 209 in the direction of arrow A at a speed V corresponding to the copy magnification, thereby scanning the entire surface of the original on the original glass 208. With the scanning of the exposure lamp 201 and the first mirror 2021, the second mirror 2022 and the third mirror 2023 also move in the direction of the arrow A at the speed V / 2. Exposure lamp 2
The position 01 is the amount of movement from the home position, that is, the number of steps of the scan motor 209 and the scan home sensor 21.
It is calculated and controlled by the 0 detection signal.

The reflected light of the document incident on the CCD sensor 204 is converted into an electric signal in the CCD sensor 204, and is subjected to analog processing, AD conversion, digital image processing, and the like in an image processing unit 205. And the image forming unit 300. Original glass 20
In addition to the original reading position on the original 8, a white shading correction plate 206 is arranged. Prior to reading image information on the original, the shading correction plate 206 is used to generate correction data for shading correction. read.

Next, the image forming section 300 will be described. First, exposure and imaging will be described. The image data sent from the image reading unit 200 or the external interface 207 is converted into printing data of each color of C (cyan), M (magenta), Y (yellow), and K (black), It is sent to the control unit of the head. Each exposure head control unit emits a laser according to the pixel value of the sent image data. And
The emitted laser light is one-dimensionally scanned by a polygon mirror 301, and each imaging unit 302C, 302
The surfaces of the photoconductors in M, 302Y and 302K are exposed.

Each of the imaging units 302C-302
In K, elements necessary for performing an electrophotographic process centering on a photoreceptor are arranged, and C, M,
By rotating each photoconductor for Y and K clockwise,
The electrophotographic process is performed continuously. Each of the imaging units 302C to 302K required for image formation is integrated for each color, and has a structure that is detachable from the main body. A latent image formed by the above-described exposure on the surface of the photoconductor in each of the imaging units 302C to 302K is
Each color is developed by a developing device. The toner images formed on the surface of the photoconductor formed by development are transferred to transfer chargers 303 </ b> C to 303 </ b> C arranged in the paper transport belt 304 so as to face the photoconductor.
At 03K, the image is transferred to a recording sheet conveyed on the sheet conveying belt 304.

Next, the feeding, conveyance, and fixing of a recording sheet will be described. The recording sheet to be transferred is supplied to the transfer position in the following order, and an image is formed thereon. Recording sheets of various sizes are set in the sheet cassettes 310a to 310c, and recording sheets of a desired size are transported by sheet feeding rollers 312a to 312c attached to the sheet cassettes 310a to 310c. Supplied to

The recording sheet supplied to the transport path is transported onto a paper transport belt 304 by a transport roller pair 313. Here, the timing sensor 306 detects a reference mark on the sheet conveyance belt 304, and adjusts the conveyance timing of the conveyed recording sheet. At the most downstream of the imaging units 302C to 302K in the recording sheet conveyance direction, three registration correction sensors 312 are arranged along the main scanning direction.
When a resist pattern is formed on
12 detects the amount of color misregistration in the main scanning direction and the sub-scanning direction of the image of each color of C, M, Y, and K, and performs the drawing position correction and the image distortion correction in the print image control unit (PIC unit). This prevents color shift on the recording sheet. Then, the transferred toner image on the recording sheet is heated and melted by the fixing roller pair 307 to be fixed on the recording sheet, and then discharged onto the discharge tray 311.

In the case of double-sided copying, the recording sheet on which the toner image is fixed by the fixing roller pair 307 is used to form an image on the back surface of the recording sheet.
9 and is guided by the duplex unit 308 to be fed again to the transport path. The paper transport belt 304 is moved up and down by a belt retracting roller 305 to move the C, M, and Y imaging units 302.
C, 302M, and 302Y can be evacuated, and the paper transport belt 304 and the photoconductor of each color can be brought into a non-contact state. That is, when a monochrome image is formed, the driving of each of the imaging units 302C, 302M, and 302Y can be stopped, so that abrasion of the photoconductor and the like can be prevented.

(2) Configuration of Image Processing Unit 205 Next, the image processing unit 205 provided in the image reading unit 200
The content of the signal processing will be described. FIG. 2 and FIG.
FIG. 3 is a functional block diagram illustrating a configuration of an image processing unit 205.
The CCD sensor 204 shown in FIG. 2 converts an original image into electrical signals separated into R, G, and B colors according to the intensity of light reflected from the original surface. The reading resolution of the CCD sensor 204 is 400 dpi, 600 dpi, 800d
Pi, 1200 dpi, etc.
The A / D conversion unit 401 receives the CCD sensor 2 based on the timing signal output from the reference drive pulse generation unit 411.
The analog signal output from the converter 04 is converted into digital data of 8 bits, that is, 256 gradations for each color information of R, G, and B.

In the shading correction unit 402, R,
Correction for eliminating unevenness in the amount of light in the main scanning direction of the image data of each color of G and B is performed. For shading correction, the shading correction plate 206 is independently provided for each color.
Are stored as reference data in an internal shading memory. Specifically, the correction can be performed by reciprocal conversion of the reference data and multiplication with the image data when scanning the document.

The line-to-line correction unit 403 converts the image data of each color in line units using an internal field memory according to the scanning speed in order to match the reading position of each of the R, G, and B sensor chips in the scanning direction. To control the delay. Due to the chromatic aberration phenomenon caused by the optical lens, the reading phase difference of each color of R, G, and B becomes larger toward the document end on the main scanning side. Due to this effect, erroneous determination may be caused by ACS determination, which will be described later, other than simple color shift. Therefore, the chromatic aberration correction unit 404 uses R,
The phase difference between G and B is corrected based on the saturation information.

In the scaling / movement control unit 406, R, G, B
By using two scaling line memories for each image data of each color, input and output are alternately operated for each line, and the writing / reading timing is controlled independently, so that scaling in the main scanning direction can be performed. Perform a move process. That is, reduction is performed by thinning out data at the time of writing to the memory, and expansion is performed by padding the data at the time of reading from the memory. In this control, interpolation processing is performed on the reduction side before writing to the memory and on the enlargement side after reading out the memory in accordance with the magnification to prevent image loss and rattling. By combining the control on this block and the scan control, not only enlargement and reduction, but also processing such as centering, image repeat, and binding margin reduction are performed.

The histogram generation unit 412 and the automatic color selection (ACS) determination unit 413 create brightness data from R, G, and B color image data obtained by preliminary scanning prior to the operation of copying a document. While the histogram is created in the memory, it is determined whether or not each dot is a color dot based on the saturation data.
The number of color dots for each two-dot square mesh is created on the memory. Based on this result, automatic copy base level control (AE processing) and automatic color selection (ACS processing) for color copy operation or monochrome copy operation are performed.

The line buffer unit 414 stores the image data of each color of R, G, and B read by the image reading unit 200 in one line.
It has a memory that can store the data for each line, and can monitor image data for image analysis for automatic sensitivity correction and automatic clamp correction of the CCD sensor 204 in the AD conversion unit 401. The HVC conversion unit 421 performs brightness (V data) by a 3 * 3 matrix operation from the data of each color of R, G, and B input via the data selector 422.
And color difference signals (Cr, Cb data).

Next, the AE processing unit 423 corrects the V data based on the background level control value described above, and outputs the Cr and Cb data according to the saturation level and the hue level set on the operation panel. Make corrections. Then, reverse HVC
The conversion unit 424 performs a 3 * 3 inverse matrix operation,
The data is reconverted into data of each color of R, G, B. In the color correction unit 430 shown in FIG.
After converting the data of each color B into density data (DR, DG, DB data), the black amount extraction unit 432 outputs
The minimum color level of the G and DB data is detected as the lower color component of the document, and at the same time, the gradation level difference between the maximum color and the minimum color of each of the R, G, and B colors is detected as document saturation data.

The DR, DG, and DB data are subjected to 3 * 6 non-linear matrix operation processing by the masking operation unit 433, and color data (C, M,
(Y, K data). A base color removal / black addition processing unit (UCR / BP processing unit) 434 converts the under color component (Min (R, G, B)) of the original into a UCR / BP corresponding to the original saturation data. The coefficient is calculated, the UCR / BP amount is determined by a multiplication process, and C,
By subtracting the undercolor removal amount (UCR) from the M and Y data,
Calculate C, M, Y data and K data. Further, the monochrome data generation unit 435 creates a lightness component from the data of each color of R, G, and B, performs LOG correction, and outputs black data (DV data). Finally, the color data selection unit 4
At 36, C, M, Y, K data as color copy images and DV data (C, M, Y) as monochrome copy images
Is white).

The area discriminating section 440 is connected to the data selector 42
Based on the image data of each color of R, G, and B input through the interface 2, an edge portion having a relatively large density difference (intensity difference) from a background image such as a character image or a line drawing such as a ruled line is determined for each pixel. An area determination signal indicating whether the pixel is an edge area pixel of an image (hereinafter, referred to as a “character image”) or a pixel existing in a halftone dot area; S10 to S13 are output. Area determination unit 44
The detailed configuration of 0 will be described later.

The image correction unit 460 performs a first smoothing process, a second smoothing process, and an edge enhancement process on each of the C, M, Y, and K data from the color correction unit 430, and performs an area determination unit. Based on the area discrimination signal output from 440, outputs a signal subjected to the first smoothing processing, outputs a signal subjected to the second smoothing processing, outputs a signal subjected to the edge enhancement processing,
Output without performing any processing is selected, and the one subjected to the selected correction processing is transferred to the print image control I / F (interface) 465.

Here, the first smoothing process is a smoothing process conventionally performed on a halftone image,
The second smoothing process is a characteristic process of the present invention, and performs a smoothing process using a filter having a different configuration from the smoothing filter used in the first smoothing process. The configuration of the smoothing filter of the first and second smoothing processes and what correction processing is performed when an area discrimination signal is output will be described later in detail. Further, the image correction unit 460 sends the data of C, M, Y, and K to the image interface unit 467 via the data selector 466 shown in FIG.

The image interface section 467 is a section for inputting and outputting image data to and from an external device. The image interface unit 467 enables simultaneous input and output of data of each color of R, G, and B, and sequential input and output of data of C, M, Y, and K. The external device can use the scanner function and the printer function of the copier 1. (3) Configuration of Area Determining Unit 440 FIG. 4 is a diagram illustrating a configuration of the area determining unit 440.

The area discriminating section 440 discriminates, from the data of each of the colors R, G, and B, what kind of area the target pixel for area discrimination (hereinafter, simply referred to as “target pixel”) exists. Then, the area determination signals S10 to S13 are transmitted to the image correction unit 4
Output to 60. The area discriminating section 440 has a lightness / chroma detecting section 441, a halftone dot detecting section 444, an edge detecting section 447, and an edge peripheral area detecting section 448. The output of each section is processed by a subsequent logic circuit. Output region determination signals S10 to S13. Hereinafter, the processing contents of each unit will be described in detail.

The lightness / saturation detecting unit 441 performs Lab conversion of the data (reflected light data) of each of the R, G, and B colors to generate chroma data and lightness (L) data S1. In the present embodiment, a detailed description of the saturation data is omitted. The halftone dot detecting section 444 includes an isolated point detecting section 442 and an isolated point counting section 443. The isolated point detecting section 442 firstly sets a window of a predetermined size centered on the target pixel, for example, 5 pixels * 5 pixels. After setting, the target pixel transmitted from the lightness / saturation detection unit 441 is compared with the lightness (L) data S1 of the surrounding pixels, and based on the comparison result, whether the target pixel is an isolated point is determined. And outputs a signal S2 indicating the result of the determination.

The isolated point counting section 443 sets a window of a predetermined size centering on the target pixel, and, based on the output signal S2 of the isolated point detecting section 442, an isolated point within the window, here a black isolated The number of points (isolated points in the case where pixels with low lightness are isolated with respect to pixels with high lightness as background) is counted. The number of counted black isolated points is compared with a preset threshold value REF1, and if the number of isolated points is larger than the threshold value REF1, the area determination signal S10 is set to high; otherwise, S10 is set to low. S10
Is high, it indicates that the target pixel is a pixel in the dot area. The configuration of the dot detection unit 444 is described in detail in, for example, JP-A-2000-59616.

On the other hand, the edge detecting section 447 comprises an edge amount calculating section 445 and an edge counting section 446. The edge amount calculating section 445 sets a window of a predetermined size centering on the target pixel, and sets the brightness / color. The edge amount is calculated from the lightness (L) data S1 generated by the degree detection unit 441 by primary differentiation or secondary differentiation. Then, the calculated edge amount is compared with a predetermined threshold value REF2 set in advance, and if the edge amount is larger than REF2, the target pixel is located in an edge area of a character image (hereinafter, referred to as a “character edge area”). Output signal (edge signal) S
3 is set to high, and otherwise, the edge signal S3 is set to low because it is not within the edge area.

The edge counting section 446 sets a window of a predetermined size around the target pixel, for example, 5 pixels × 5 pixels, and based on the edge signal S 3 from the edge amount calculation section 445, sets the window within the window. The number of pixels determined to be within the existing character edge area is counted. Then, it is determined whether or not the value of the counted result is greater than a predetermined threshold value REF3. If it is determined that the value is greater, the target pixel is determined to be a pixel in the character edge area, and the output signal S13 is set to high. When it is determined that the pixel is not a pixel in the character edge area, the area determination signal S13 is set to low (hereinafter, a pixel determined to be a pixel in the character edge area is referred to as an “edge pixel”).

The reason why the number of edge pixels existing in the window is counted to determine whether the number of edge pixels is larger than the predetermined threshold value REF3 is as follows. That is, in the character edge region, there is a high possibility that edge pixels are continuously present, and in that case, a predetermined number or more of edge pixels should be included in the window. Therefore, the value of the counted result is equal to the threshold R
If EF3 or less, erroneous determination is prevented as much as possible because there is a high possibility that the image does not correspond to the character edge region, and there is also a possibility that it corresponds to a pixel in the halftone dot region that is in contact with the character image It is because it is preferable to do. Considering that it is highly likely to be continuous, a condition for determining whether or not the number of continuous edge pixels is a predetermined value or more is added to the condition for determining whether or not the pixel is in the character edge area. By doing so, more accurate determination can be realized.

Further, in the present embodiment, in order to make it easier to determine the edge area of the character image existing in the halftone dot area (hereinafter referred to as "character edge area in halftone dot"), a plain (white) background is used. The value of the REF2 is set to be smaller than a threshold value usually used for determining a character edge region when a character image exists. This is because, in the character edge region in the halftone dot, since the character edge exists on the background of the halftone dot, the edge amount is lower than that in the case of the plain color, so the threshold is set small and the edge is set. This is because it is preferable to make the detection easy.

The edge peripheral area detecting section 448 sets a window of a predetermined size centered on the target pixel, in this embodiment, a size of 3 pixels vertically × 3 pixels horizontally, and sets a window other than the target pixel in the window. Among the other pixels, it is determined whether or not an edge pixel exists.
The target pixel is determined to be a pixel (hereinafter, referred to as “edge peripheral pixel”) existing in a region around the edge pixel (hereinafter, referred to as “edge peripheral region”), and the output signal S is determined.
Let 21 be high. On the other hand, when it is determined that there is no edge pixel, S21 is set to low. The operation of the edge peripheral area detecting section 448 will be specifically described with reference to FIG.

FIG. 5 is a schematic diagram showing a part of the edge of the character image existing in the halftone dot area in a range of 5 pixels vertically × 5 pixels horizontally. Here, V11, V21... V5
It is assumed that 10 pixels 1 and V12, V22,..., V52 are pixels in the character image area and are determined to be edge pixels, and the other 15 pixels are pixels belonging to the dot area.

In such an image, for example, when the target pixel is V33, three edge pixels V22, V32, and V42 are set in a window having a size of three vertical pixels * three horizontal pixels.
Is present, V33 is determined to be an edge peripheral pixel. On the other hand, assuming that V34 is the target pixel, no edge pixel is present in the window, so that it is determined that the pixel is not an edge peripheral pixel.
Thus, in the image of FIG.
V13, V23,... Are determined as edge peripheral pixels.
This is the case when V53 has become the target pixel.

Returning to FIG. 4, when both the area discrimination signals S10 and S13 are high, the AND circuit 449 outputs
The area determination signal S11 is set to high. That is, when S11 is high, the target pixel is determined to be an edge pixel of the character image existing in the halftone dot image.
The AND circuit 451 outputs the output signals S21 and S22 (S
When both of the signals 13 are inverted by the inverter 450) are high, the area determination signal S12 is set to high. That is, when S12 is high, the target pixel is determined to be an edge peripheral pixel. Note that S2
The logical sum of the signal of 1 and S22 is obtained for the following reason. That is, in the configuration of FIG. 5, for example, in the configuration of FIG. 5, even if a pixel is determined to be an edge peripheral pixel by the edge peripheral region detection unit 448, the pixel is determined to be an edge pixel. V
This is because when the target pixel is 32, V32 is also determined as an edge peripheral pixel, and as a result, it may not be possible to distinguish between the edge peripheral pixel and the edge pixel. Therefore, in the present embodiment,
If the target pixel is determined not to be an edge pixel but to be a peripheral pixel by taking a logical sum, the region determination signal S
12 is set to be high, so that the edge pixel and the edge peripheral pixel are distinguished.

(4) Processing Contents of Image Correction Unit 460 FIG. 6 is a diagram showing the configuration of the image correction unit 460. As shown in the figure, the image correction unit 460 includes smoothing processing units 461 and 462, an edge enhancement processing unit 463, and a selector 464. The image correction unit 460 receives the respective density data of C, M, Y, and K from the color correction unit 430, and the data is directly input to the selector 464 by the 0, 1,.
The signals are input to the terminals Nos. 2, 3, and 7, and are also input to the smoothing processing units 461 and 462 and the edge enhancement processing unit 463.

The smoothing processing unit 461 uses a well-known smoothing filter conventionally used for smoothing a halftone dot image, for example, a filter of 3 × 3 pixels as shown in FIG. Then, a smoothing process is performed on the density data of the target pixel by performing a moving average by weighted addition of the density values of the peripheral pixels, and the selector 464.
Send to The “first smoothing process” is a smoothing process performed by the smoothing processing unit 461.

On the other hand, the smoothing processing section 462 separately performs smoothing processing on the density data of the target pixel using not only one filter but also a plurality of filters, here eight filters (see FIG. 9). The signal subjected to the smoothing processing using one filter satisfying a predetermined condition is sent to the selector 464. The “second smoothing process” is a smoothing process performed by the present smoothing processing unit 462. The configuration of the smoothing processing unit 462 will be described later.

The edge emphasis processing section 463 performs an edge emphasis process on the density data of the target pixel by using a known filter conventionally used for the edge emphasis process, and sends the data to the selector 464. The selector 464 receives the area determination signal S10 sent from the area determination unit 440,
According to the combination of S11 and S12, a signal to be output from one output terminal is selected from the signals input to the input terminals 0 to 7, and the selected signal is printed image control I / F 465. Output to

The following (Table 1) summarizes the processing contents of the selector 464.

[0049]

[Table 1]

As shown in the above (Table 1), the image correction unit 460 of the present embodiment determines that the target pixel is a pixel in the halftone dot region and a pixel existing in a region other than the edge peripheral region. (When the area determination signal S10 is high, S11 is low, and S12 is low), the signal input to the input terminal 4 is output. That is, the smoothing processing unit 4
The data that has been subjected to the first smoothing process by 61 will be output.

On the other hand, when it is determined that the target pixel is a pixel existing in an edge peripheral area in the halftone dot area (hereinafter, referred to as "character edge peripheral area in halftone dot") (the area determination signal S10 is high and S11 is high). Low, when S12 is high),
The signal input to the input terminal 5 is output. That is, the data subjected to the second smoothing processing by the smoothing processing unit 462 is output.

When the target pixel is determined to be a pixel in the character edge area in the halftone dot (the area determination signal S10 is high,
When S11 is high and S12 is low), the signal input to the input terminal 6 is output. That is, the data subjected to the edge enhancement processing by the edge enhancement processing unit 463 is output. In addition, the pixels in the region other than the above,
For example, in the present embodiment, pixels which are neither in the halftone dot region nor in the character edge region are output without any processing (through).

That is, in this embodiment, the smoothing processing section 46 applies the density data to each pixel.
1, the first smoothing process, the second smoothing process by the smoothing processing unit 462, and the edge enhancement process by the edge enhancement processing unit 463 are performed. Print image control I / F
The signal S sent from the area discriminating unit 440 is sent to the
Only those selected by the selector 464 based on 10 to S12.

FIG. 8 is a diagram showing the configuration of the smoothing processing unit 462. As shown in the drawing, the smoothing processing unit 462 includes a line buffer 470, a filter processing unit 4
71 to 478, direction detection circuit 479 and selector 48
Consists of zero. The line buffer 470 temporarily accumulates data for the lines required for the filter processing in the filter processing units 471 to 478.

The filter processing unit 471 performs smoothing processing on the density of the target pixel by performing a moving average by weighted addition of the density values of peripheral pixels using a smoothing filter 481 of 3 * 3 pixels shown in FIG. Is a circuit for outputting the obtained density data. In particular,
In the case of an image composed of a matrix of 3 vertical pixels × 3 horizontal pixels, the i-th row and the j-th column (1 ≦ i ≦ 3, 1 ≦ j ≦
When the density data of the pixel Vij in 3) is Dij, and similarly, the coefficient of the pixel in the i-th row and the j-th column of the smoothing filter 481 is Kij, the density data D22 of the target pixel V22 is D22 = Σ (Dij * Kij) / Σ (Kij) (Equation 1) is calculated and smoothed. For example, when the target pixel is V33 in the image shown in FIG.
When the filter 481 is used, the density data D33 of V33
Is D33 = (D23 * 2 + D24 * 2 + D33 * 4 +
D34 * 2 + D43 * 2 + D44 * 2) / (2 + 2 + 4
+ 2 + 2 + 2).

Similarly, filter processing units 472 to 478
Output density data smoothed using the smoothing filters 482 to 488, respectively. As a result, eight pieces of smoothed data are generated for one piece of data and output to the selector 480.
Note that each of the smoothing filters 481 to 488 includes three pixels whose coefficient (weighting amount) becomes “0” when filtering around the center pixel serving as the target pixel. The position of the coefficient “0” is set so that the weighting direction changes in eight directions, left, right and upper right, lower right, lower left, and upper left. The reason for such a configuration will be described later.

Next, the configuration of the direction detection circuit 479 is shown in FIG.
This will be described with reference to FIG. As shown in FIG. 10, the direction detection circuit 479 includes a line buffer 490 and a filter processing unit 49.
1 to 498, and a MAX selection circuit 499, and detects in which direction a row formed by continuation of edge pixels (hereinafter, referred to as an “edge pixel row”) is positioned with respect to the target pixel.

The line buffer 490 temporarily accumulates data for the lines necessary for the filter processing in the filter processing sections 491 to 498. Filter processing unit 491
Is a filter 501 of 5 × 5 pixels shown in FIG.
Is used to filter the density data of the target pixel. Similarly, the filter processing units 492 to 498 perform filtering using the filters 502 to 508.
The method of operation by filtering is the same as that of the smoothing filter 481 and the like. That is, when the range of the variables i and j is 1 ≦ i ≦ 5 and 1 ≦ j ≦ 5, and the target pixel is V33, the density data D
33 is D33 = Σ (Dij * Kij) / Σ (Kij)
Becomes

Here, each of the filters 501 to 508 performs filtering when the edge pixel row is in contact with the target pixel and the edge pixel row matches the pixel row having the coefficient “1”. The position of the pixel having the coefficient “1” is set such that the calculated value of “1” becomes the largest. Therefore, for example, when the target pixel is set to V33 in the image shown in FIG.
When the processing is performed by 98, the calculated value when the filter 501 is used becomes the largest. Similarly, if the target pixel is V
Also in the case of 23, the calculated value when the filter 501 is used becomes the largest. Further, for the target pixel V33,
If the edge pixels are V14, V23, V32, and V41, a filter 507 in which a coefficient “1” is set to a pixel having the same positional relationship as each of the pixels V14, V23, V32, and V41 is used. The calculated value is the largest.

That is, when the target pixel is in contact with the edge pixel column, the coefficient set in the filter of the filter processing unit having the largest value among the values calculated by the respective filter processing units 491 to 498. The pixel row of “1” substantially represents the position of the edge pixel row with respect to the target pixel. Hereinafter, filters 501 to 508
, The position of each edge pixel column with respect to the target pixel is represented by a filter 501... 5
08 to the first to eighth positions.

The MAX selection circuit 499 determines which of the output values from the filter processing units 491 to 498 is the filter processing unit having the maximum value, and from the result of the determination, determines the edge pixel row for the target pixel. And a signal indicating the position, for example, “1” for the first position,
If the position is the second position, the signal indicating "2" if the position is the eighth position, the signal indicating "8" is converted into a 3-bit signal, and the selector 48
To output to 0.

Returning to FIG. 8, the selector 480 receives a signal from the direction detection circuit 479. When the signal is, for example, “1”, that is, when the position of the edge pixel row with respect to the target pixel is the first position. Selects a signal from the filter processing unit 471 and outputs the signal to the selector 464. If "2", the signal from the filter processing unit 472 is selected. Similarly, if "3"... "8", the signal from the filter processing unit 473.
64.

That is, the selector 480 performs smoothing using a filter in which a pixel having a coefficient “0” is located at a position specified as an edge pixel row with respect to the target pixel by the direction detection circuit 479 among the smoothing filters 471 to 478. You will select the processed data. This configuration is such that, when performing smoothing processing on a pixel belonging to a character edge peripheral area in a halftone dot as a target pixel, the density data of a pixel belonging to a halftone dot character edge area adjacent to the target pixel is used. This is because the reference is not referred to and the edge is emphasized more than before.

That is, for example, in the case of the smoothing filter 481, the coefficient corresponding to each pixel located on the left, upper left, and lower left adjacent to the target pixel is “0”.
Therefore, in the image shown in FIG.
The density data before the smoothing process of V55 is, for example, as shown in FIG. 12 (the higher the numerical value, the higher the density.), And the pixel V33 (the hatched pixel) is the character in the halftone dot. If it is determined that the pixels V22, V32, and V42 are pixels in the character edge area in the halftone dot, the pixel V22, V32, and V42 are pixels in the edge peripheral area. The density data "255" of the pixels V22, V32, and V42 in the character edge area in the dot is not referred to at all, and the density data D33 after smoothing is "72.9" from the above equation.

On the other hand, conventionally, a smoothing filter as shown in FIG. 7 is used uniformly for pixels in the halftone dot region regardless of whether they are pixels in the edge peripheral region. Is used as a target pixel, the density data of the three pixels V22, V32, and V42 are also reflected in the calculation result, and the density data D33 after smoothing is larger than that when the smoothing filter 481 is used according to the above equation. ".

That is, if the smoothing filter 481 of the present embodiment is used, the density difference between the density data D33 of the pixel V33 after smoothing and the pixel V32 of the character edge area in the adjacent halftone dot becomes larger than in the related art. The fact that the difference in density from the pixel V32 becomes large can be said to be the same as the emphasis of the edge, so that when the character image is viewed, the edge appears sharper than before. Become.

This is the same for the pixels in the area around the character edge in other halftone dots. For example, V23
Is a target pixel, in the present embodiment, a signal indicating the first position of the edge pixel row is sent to the selector 480 by the direction detection circuit 479 as described above, so that smoothing was performed using the smoothing filter 481. Data will be output. In this case, according to the above equation, the pixels V12, V22, V
32 is not referred to, and the density data D23 after smoothing is "7" as in the case of V33.
2.9 ”. On the other hand, conventionally, each pixel V12, V2
2. The density data of V32 is referred to and becomes "128".

FIG. 13 shows each of the pixels V11 to V5 shown in FIG.
In the case where the density data before smoothing shown in FIG. 5 is shown in FIG. 12, V23 in the halftone dot region is different between the case of the present invention in which the density data of the pixel in the character edge region in the halftone dot is not referred and the conventional case of reference. , V33, V43 and V2
FIG. 4 is a diagram specifically showing density data after smoothing when each of V, V34, and V44 is a target pixel.

FIG. 13A shows the case of the prior art, and FIG. 13B shows the case of the present invention. FIG. 13B shows density data of pixels V24, V34, and V44 when filtering is performed using a conventional smoothing filter. this is,
In the present embodiment, as described with reference to FIG. 6 described above, when the target pixel is a pixel in the halftone dot region and is a pixel other than the edge peripheral pixel, the first smoothing process by the smoothing processing unit 461 is performed. This is because the applied data is output from the selector 464.

FIG. 14A is a diagram showing the density values of the respective pixels shown in FIG. 12 in the form of a bar graph, and FIG.
FIG. 14C is a diagram showing a bar graph of the density value of each pixel shown in FIG. 13A after smoothing, and FIG.
FIG. 7 is a bar graph showing density values after smoothing of each pixel shown in FIG. 13 (a) and 13 (b) and FIGS. 14 (b) and 14 (c), the density data D23, D43 of the pixels V23, V33, V43 in the area around the character edge in the halftone dot.
As is clear from the comparison between the conventional case and the case of the present invention, 33 and D43 are lower in the present invention than in the conventional case, and adjacent pixels V24, V34 and V44 in the halftone dot region.
Is almost equal to the concentration. That is, as the density difference from the pixels in the character edge region in the halftone dot increases, the edges appear more sharp (sharp), and adjacent pixels V24, V34, V4 in the halftone dot region.
4 is smaller than that in the related art, so that an effect of making the image look smoother can be obtained.

(5) Modifications (5-1) In the above embodiment, as the smoothing filters 481 to 488 in the second smoothing processing, those having a size of 3 pixels vertically × 3 pixels horizontally are used. The size (size) of the filter is not limited to this, and is determined in consideration of the resolution of the image, the strength of the smoothing process, and the like. For example, in the case of a size of 5 vertical pixels × 5 horizontal pixels, the position of the coefficient “0” is V11 ·· V51 and V12 ·· V located on the left side of the center pixel V33 if it is a filter for the smoothing filter 481.
52, and a smoothing filter 48
8 for V25, V34, V4
3, V52, V35, V44, V53, V45, V5
4, V55. That is, similarly to the above embodiment, “0” is set at a position where the density data of the edge pixel is not referred to.

Note that the smoothing filters 481-48
8 is 5 pixels long × 5 pixels wide, even if the size of the window used for detecting the edge surrounding pixels in the edge surrounding area detection unit 448 is 3 pixels long × 3 pixels wide. If so, the smoothing filter of 5 vertical pixels × 5 horizontal pixels is used only when the pixels adjacent to the edge pixels (V13, V23,..., V53 in FIG. 5) are the target pixels.

According to the size of the smoothing filters 481 to 488 being 5 pixels * 5 pixels, the size of the window can be expanded to 5 pixels * 5 pixels. In this case, as the window becomes larger, the edge peripheral area becomes wider, and a width range of two pixels surrounding the periphery of the edge pixel, for example, V13, V13 in FIG.
V54, V24,..., V54 in addition to 23... V53 are also determined as edge peripheral pixels. Therefore, when these pixels become target pixels, the pixels are subjected to a second smoothing process using a smoothing filter having a size of 5 vertical pixels × 5 horizontal pixels.

As described above, the smoothing filter having the size of 5 vertical pixels × 5 horizontal pixels also has 3 vertical pixels × 3 horizontal pixels.
Since the position of the coefficient “0” is set so as not to refer to the density data of the edge pixel as in the case of the pixel, the density value when each pixel determined as an edge peripheral pixel becomes the target pixel is Are smoothed so as to have almost the same density values as the pixels in the halftone dot region.

The smoothing filters 481 to 48
The coefficient other than “0” set to 8 is not limited to the above value, and is determined in consideration of the resolution and the like. (5-2) In the above embodiment, when the edge peripheral pixel is the target pixel, the smoothing process is performed without referring to the density of the edge pixel.
488 is set so that the coefficients of three pixels among the eight peripheral pixels are “0”, but the coefficients are not limited to “0”. To further emphasize the edge,
In order to set the coefficient of one pixel to, for example, “−1” or to further smooth the density of each pixel in the halftone dot area, the coefficient of the three pixels is set to, for example, “0.1”.
This is because it may be more convenient in terms of image quality to set. Further, the coefficients of the three pixels may be set differently.

That is, when the pixel in the halftone dot region is subjected to the smoothing process as a target pixel, a pixel determined to be an edge pixel is included in pixels within a range of 3 * 3 pixels including the target pixel. In some cases, if the coefficient (weighting amount) for the edge pixel is set to be smaller than the coefficient when the edge pixel is determined to be a pixel in the halftone dot region, the calculation is performed at least in the conventional case. Can be obtained.

When the size of the smoothing filter is increased, for example, if a pixel adjacent to the edge pixel in the peripheral area of the edge is set as a target pixel, a solid portion of the character image is within a range to be filtered by the filter. May be included. In such a case, it is configured to determine whether or not the pixels within the range belong to the character image (including the edge pixels) region, and determine that the pixel is within the character image region within the range. When there is a pixel that has been set, if the coefficient for the pixel is set to be smaller than the coefficient when the pixel is determined to be a pixel in the halftone dot region, the same effect as described above can be obtained. Become.

Here, the determination as to whether or not a pixel is within the character image area can be made by the edge detection section 447 for the character edge area. For the solid region, for example, the pixel to be determined is not an edge pixel and is not an isolated point, and the density of the pixel and all pixels belonging to a predetermined range centered on the pixel is a predetermined density (for example, print density). (The density of the background of the paper to be performed).

(5-3) In the above embodiment, the case where a character image is drawn in a halftone image has been described.
The present invention is not limited to this case, and can be applied to, for example, a case where a character image is drawn in a density flat region where a density change is small. Normally, the flat density area is subjected to smoothing processing for each pixel in the area in order to smooth the image. In this case, the edge of the character image is substantially enhanced by applying the present invention. Because it can be. That is, the present invention can be generally applied to a case where a character image is drawn in an area to be subjected to smoothing processing such as a halftone dot image or a flat density area. When there is a pixel determined to be in the character image area, the weighting amount for the pixel is determined assuming that the pixel is not a pixel in the character image area, that is, a pixel to be subjected to smoothing processing such as a halftone dot area. If it is smaller than this, the effect of the edge enhancement can be obtained. In addition,
When the present invention is applied to a flat density region, a flat density region determining unit that determines whether the target pixel exists in the flat density region is arranged in the region determining unit 440 instead of the halftone dot detecting unit 444. Judgment as to whether or not the area is a flat density area is described in
Since the details are described in, for example, Japanese Patent No. 38261, description thereof is omitted here.

(5-4) In the above embodiment, the density data of each pixel is weighted by each of the smoothing filters 481 to 488. For example, R, G, and B of each pixel are weighted.
It is also possible to weight the brightness data by color and to obtain the density data of C, M, Y, K colors based on the data. (5-5) In the image correction unit 460 of the above embodiment, the smoothing processing unit 46 performs processing on the density data of one pixel.
1, 462 and the respective processes by the edge enhancement processing unit 463 are performed in parallel, and data to be output is selected based on the determination result by the area determination unit 440. For example, based on the determination result by the area determination unit 440, Then, a processing unit to be operated may be selected. This means
Filter processing section 47 in smoothing processing section 462
The same applies to the smoothing processing by 1 to 478. That is, the filter processing unit to be operated may be selected based on the detection result from the direction detection circuit 479.

[0081]

As described above, according to the image processing apparatus of the present invention, it is determined that the target pixel is not within the character image area, and that the target pixel is determined to be within the character image area within a predetermined range including the target pixel. When a pixel exists, the weighting amount for the pixel is set to be smaller than the weighting amount when the pixel is not included in the character image area, so that the density of the target pixel after the smoothing process is lower than before. ,
The edge is more emphasized than in the related art, so that there is an effect that deterioration in image quality can be suppressed.

[Brief description of the drawings]

FIG. 1 is a schematic cross-sectional view illustrating an entire configuration of a copying machine 1.

FIG. 2 is a functional block diagram illustrating a configuration of an image processing unit 205.

FIG. 3 is a functional block diagram illustrating a configuration of an image processing unit 205.

FIG. 4 is a diagram showing a configuration of an area determination unit 440.

FIG. 5 is a schematic diagram showing a part of an edge of a character image existing in a halftone dot area, which is decomposed into a range of 5 pixels × 5 pixels.

FIG. 6 is a diagram illustrating a configuration of an image correction unit 460.

FIG. 7 is a diagram illustrating a configuration example of a conventional smoothing filter.

FIG. 8 is a diagram showing a configuration of a smoothing processing unit 462.

FIG. 9 is a diagram illustrating a configuration example of smoothing filters 481 to 488 of 3 pixels * 3 pixels used in a smoothing processing unit 462.

FIG. 10 is a diagram showing a configuration of a direction detection circuit 479.

11 is a diagram illustrating a configuration example of filters 501 to 508 used in a direction detection circuit 479. FIG.

12 is a diagram illustrating density data of pixels V11 to V55 illustrated in FIG. 5 before smoothing processing is performed.

13A is a diagram illustrating pixels V11 to V55 illustrated in FIG.
FIG. 13B is a diagram showing a density value when each pixel in the halftone dot region is smoothed using a conventional smoothing filter in a case where the density data before smoothing is shown in FIG. It is a figure which shows the density value at the time of using the smoothing filter of this invention.

14 (a) is a graph showing the density value of each pixel shown in FIG. 12 in a bar graph, and FIG. 14 (b) is a graph showing the density value of each pixel shown in FIG. 13 (a) after smoothing. FIG. 13C is a diagram showing a bar graph of the density value of each pixel shown in FIG. 13B after smoothing.

FIG. 15 is a diagram in which a part of an image used for explaining a conventional smoothing process is decomposed into a range of 5 pixels * 5 pixels.

[Explanation of symbols]

 200 Image reading unit 205 Image processing unit 440 Area discriminating unit 441 Brightness / chroma detecting unit 442 Isolated point detecting unit 443 Isolated point counting unit 444 Halftone dot detecting unit 445 Edge amount calculating unit 446 Edge counting unit 447 Edge detecting unit 448 Edge periphery Area detection section 460 Image correction section 461, 462 Smoothing processing section 463 Edge enhancement processing section 464, 480 Selector 471-478, 491-498 Filter processing section 479 Direction detection circuit 481-488 Smoothing filter 499 MAX selection circuit 501-508 Filter

Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat II (reference) H04N 1/40 H04N 1/40 F F term (reference) 2C062 AA24 2C262 AA24 AB13 BB01 BC13 DA03 DA11 EA04 EA07 EA08 5B057 AA11 BA02 CA01 CA02 CA07 CA08 CA12 CA16 CB01 CB02 CB07 CB08 CB12 CB16 CC01 CE03 CE05 CE06 CH08 5C077 LL19 MP01 MP02 MP06 MP08 PP02 PP03 PP27 PP28 PP47 PP68 PQ12 SS02 TT06 5L096 AA02 AA03 AA06 DA01 EA06 FA43 FA04 GA02

Claims (5)

[Claims] 1. An image processing apparatus comprising: a smoothing unit that weights data of pixels within a predetermined range including a target pixel and smoothes the data of the target pixel based on the weighted pixel data. And for each pixel within a predetermined range including the target pixel, a character image region determining unit that determines whether or not the pixel is present in a character image region. When the pixel is determined not to be within the image area, and a pixel determined to be within the character image area exists within the predetermined range, the weighting amount for the pixel is determined by the weighting amount when the pixel is determined not to be within the character image area. An image processing apparatus characterized in that the image processing apparatus is smaller than the image processing apparatus. 2. The image processing apparatus according to claim 1, further comprising: a halftone dot determining unit configured to determine whether the target pixel is within a halftone dot region. 2. The image processing apparatus according to claim 1, wherein a weighting amount for a pixel determined to be in the character image area is smaller than a weighting amount when the pixel is in a halftone dot area. 3. The image processing apparatus according to claim 1, wherein a weighting amount for data of a pixel determined to be within the character image area is zero. 4. A plurality of types of filters having a configuration in which a weighting amount for pixels having a predetermined positional relationship with respect to a target pixel within the predetermined range is zero, and wherein the smoothing unit includes the plurality of types of filters. The image according to any one of claims 1 to 3, wherein smoothing is performed using a filter configured such that a weighting amount for a pixel determined to be within the character image area becomes zero. Processing equipment. 5. An image forming apparatus for forming an image based on data smoothed by an image processing apparatus, wherein the image processing apparatus includes the image processing apparatus according to claim 1. An image forming apparatus comprising: