JP2000278522A - Image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processor capable of expressing natural images at the time of dividing pixels into small areas and raising a resolution. SOLUTION: A small area division part 402 divides the pixel of an edge part into plural small areas and raises the resolution of the edge part. For input image data, an edge attribute judgment part judges edge attributes (thickening pattern and thinning pattern for instance) and a gradation value to be supplied to the small area corresponding to the edge attributes is decided by taking the degradation values of the peripheral small areas into consideration in an edge smoothing processing part 406. In such a manner, by deciding the gradation value by using the small area and the peripheral gradation values, even in the case that multi-valued image data are inputted, the kind of an area is accurately discriminated and the gradation value to be outputted does not become unnatural.

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 in a copying machine, a printer, and the like.

[0002]

2. Description of the Related Art In order to improve the quality of an edge portion of an image,
A technique for increasing the resolution of an edge portion of a binary image has been proposed. For example, a binary input image is converted into a multi-valued image, and a plurality of binarization circuits binarize the multi-valued image data with different thresholds. Thereby, edges can be expressed smoothly. It is also known that when increasing the resolution of a binary image, appropriate replacement data is set by switching the replacement data in accordance with the magnification. In addition, there has been proposed a technique of determining what edge the multi-valued image has based on the gradation difference, and shifting the density center of gravity within one pixel to increase the resolution of the edge portion.

[0003]

However, the above-described edge smoothing technique is intended for binary image data.
It is not considered that multi-valued image data is input.
In order to support multi-valued image data, it is necessary to set many output gradation values according to the attributes. Further, in the method of shifting the density center of gravity in one pixel, there is a portion where the tone value becomes extremely high or low with respect to the peripheral pixels, and therefore, an image may look unnatural. For example, in the case of halftone characters, the density of the edge part becomes high and a phenomenon like a border character occurs, and in the case of the character on the density base, white spots occur near the edge part. .

An object of the present invention is to provide an image processing apparatus capable of expressing a natural image when dividing a pixel into small areas and increasing the resolution.

[0005]

According to the present invention, there is provided an image forming apparatus comprising: a dividing section for dividing each pixel into a plurality of small areas in input image data; In units, the attribute determination unit that determines the edge attribute in consideration of the data of the small region of interest and its surrounding small regions, and the edge attribute determined by the attribute determination unit in units of the small region divided by the division unit. Accordingly, a gradation value setting unit for setting the gradation value of the small area from the gradation value of the small area and the surrounding small areas. Here, the dividing unit divides the pixel of the edge into a plurality of small areas, and increases the resolution of the edge to increase the resolution of the character edge and the like of the image so that the edge looks smooth. In the edge resolution increasing (smoothing) process, the attribute determination unit compares the tone values of the small area of interest and its surrounding small areas and determines the edge attribute (for example, a thickened pattern and a thinned pattern).
To determine. The gradation value is set according to the determined edge attribute. In this manner, by using the small area and the gradation values around the small area, even when multi-valued image data is input, it is possible to accurately determine what kind of area the multi-valued image data is. Even when the resolution is increased and the resolution is increased, the gradation value of each small area is determined from the surrounding gradation values. Therefore, with a simple configuration, the output gradation value maintains continuity with the peripheral region, so that an unnatural image does not occur. For example, the gradation value setting unit may select the gradation value of the small area and the maximum value and the minimum value of the gradation value of the small area and the surrounding small areas according to the edge attribute. Set the gradation value of the small area.

[0006]

Embodiments of the present invention will be described below with reference to the accompanying drawings. In the drawings, the same reference symbols indicate the same or equivalent ones. FIG. 1 shows the overall configuration of a color digital copying machine. The copying machine includes an automatic document feeder 100, an image reading unit 200, and an image forming unit 300. Normally, the automatic document feeder 100
The document conveyed to the image reading position by the
0, and reads the read image data into the image forming unit 3.
00 to form an image. The interface 207 enables connection with an external device.

Next, the image reading unit 200 will be described.
The reflected light of the original on 8 is guided to a lens 203 by a group of three mirrors 202 and forms an image on a CCD sensor 204. The exposure lamp 201 scans the document glass 2 in the direction indicated by the arrow by the scanner motor at a speed corresponding to the magnification.
08 can be scanned over the entire surface.
The reflected light of the document incident on the CCD sensor 204 is converted into an electric signal in the sensor, and the image processing circuit 205 performs analog processing, A / D conversion, and digital image processing of the electric signal. It is sent to the forming unit 300.

Next, the tandem image forming section 300 will be described. Imaging units 302c, 30
2m, 302y, and 302k are arranged vertically in a line along the sheet conveying direction of the sheet conveying belt 304. The image data sent from the image reading unit 200 or the interface unit 207 is converted into cyan (C), magenta (M), yellow (Y), and black (K) print data, and the control unit of each exposure head. (Not shown). Each exposure head control unit emits a laser in accordance with an electric signal of the image data sent thereto, and the light is one-dimensionally scanned by a polygon mirror 301, and each of the imaging units 302c, 302m, 302y, 302k.
Expose the photoconductor inside. Inside each imaging unit, elements necessary for performing an electrophotographic process centering on a photoreceptor are arranged. Each image forming process is continuously performed by rotating each of the photoconductors for C, M, Y, and K clockwise. The latent image on the photoconductor in each imaging unit is developed by each color developing unit. Sheets of a desired size are supplied to the sheet cassettes 310a, 310b, 3b.
From 10c, the feed roller 312 and the transport roller pair 313
To the paper transport belt 304. The toner images on the photoconductors are transferred to transfer chargers 303 c and 303 installed in the paper transport belt 304 so as to face the above-described photoconductors.
m, 303y, and 303k, the sheet transport belt 304
The image is transferred onto the upper sheet. Here, the timing sensor 306 detects a reference mark on the paper transport belt 304, and adjusts the transport timing of the transported paper. The transferred toner image on the sheet is heated and melted by the fixing roller pair 307 to be fixed on the sheet, and then discharged to the tray 311.

Further, at the most downstream of the imaging unit, three registration correction sensors 312 are connected to the belt 304.
Are arranged in a line in a direction (in the main scanning direction) perpendicular to the carrying direction of the image forming apparatus. When a resist pattern on the paper transport belt 304 is formed, the sensor detects the amount of color misregistration of the Y, M, C, and K images in the main and sub scanning directions, and corrects the drawing position correction and image distortion in the image data controller. By performing the correction, the color shift of the C, M, Y, and K images on the paper is prevented.

In the image data control unit shown in FIG. 2, image data from the image processing unit or the interface unit 207 of the image reading unit 200 is subjected to fixed-length compression and variable-length code conversion via an image interface unit 320. To the encoding unit 322. Specifically, the image is divided into 4 * 4 dot blocks by block truncation coding, and the data of each block is compressed to 48 bits. Next, the fixed length data is converted into 2, 10 or 50 bits by the GBTC code data LD.
Next, the frame memory unit 324 performs delay control in the sub-scanning direction, and corrects an image position corresponding to the distance between the Y, M, C, and K photoconductors. Next, the fixed-length data conversion unit 326
Converts variable-length code data into fixed-length data (GBTC compressed data). Next, the main scanning position adjustment unit 600
The dpi skew correction unit / GBTC decompression unit 328 corrects the drawing position in the main scanning direction, converts the one-sided document position reference to the center sheet position reference, and reduces the skew in the sub-scanning direction by 600d.
pi, and the GBTC compressed data is 8-bit (256
(Gradation) data. In the 600 dpi skew correction, the skew is roughly corrected for the mechanical variation of each device. Next, the tone reproduction unit 330 performs 2400 dpi resolution conversion on the skew in the sub-scanning direction, and performs edge smoothing processing, gamma correction, and screen processing. The obtained Y, M, C, K data is sent to the exposure head control unit.

FIG. 3 shows the tone reproduction section 330. 600
The image data of dpi is stored in the line memory block 400.
(See FIG. 4), and six lines of data are output in parallel. Next, in the small area dividing unit 402 (see FIG. 5), one pixel of 600 dpi is divided into four small areas,
The image data is converted into data having a resolution of 2400 dpi for 5 * 4 = 20 lines. The following processing is performed in parallel for each of the four small areas.

Next, different image processing is performed on the high-resolution data at the edge portion and the non-edge portion. For the non-edge portion, the small area division section 402 performs interpolation interpolation type resolution conversion, and the gamma correction section 404 performs gamma correction on the converted data. The processing of the non-edge part will not be further described below. For the edge portion, an edge smoothing processing unit 406 (FIG. 1)
After the edge is smoothed in (4, FIG. 15), gamma correction is performed in the gamma correction unit 408. As described in detail later, the edge smoothing processing unit 406 determines a gradation value in small area units according to the attributes (thickening pattern and thinning pattern) of the edge pixels. Here, since the gradation value is determined from the surrounding gradation values, the outputted gradation value does not become unnatural. Processing corresponding to each attribute is performed, and the data is sent to the data selection unit 410. The data selection unit 410 selects and outputs the processed image data according to the determination result of the area determination unit 412. As described above, in the edge smoothing processing, the gradation value of the small area is determined in consideration of the surrounding gradation values.
The output gradation value does not become unnatural. Conventionally, the gradation value of a small area obtained by dividing a pixel to be processed does not have continuity with the gradation value of a peripheral pixel, so that the density becomes higher or lower than that of the peripheral pixel, resulting in unnaturalness. Image was sometimes output. In the present embodiment, such a phenomenon is prevented.

FIG. 4 shows a line memory block 400. The line memory block 400 includes five line memories (FIFO memories) 4001, 4002, 4003,
4004 and 4005. The input data VI is sequentially delayed by the line memories 4001 to 4005, and six lines of data V1 to V6 are output in parallel.

FIGS. 5 to 9 show the small area dividing unit 402. FIG. One pixel V of 600 dpi is divided into a plurality (four) of small areas in the sub-scanning direction. Here, the pixel data V is converted into small area data VA, VB, VC, and VD. As shown in FIG. 5, the line memory block 400
Are input in parallel. The resolution conversion processing units 4021 to 4025 output data obtained by quadrupling the resolution of two adjacent lines of data. At the edge portion, the same data is simply assigned to the small area to increase the resolution. For example, the resolution conversion processing unit 4023 outputs the data V3 and V4 of two adjacent lines.
VA3, VB3, V obtained by converting the resolution to 4 times
C3 and VD3 are output.

In the small area dividing section 402, as shown in FIGS. 6 to 9, data of a 5 * 20 small area obtained from a pixel of a matrix of 5 rows * 5 columns including a center pixel for each same small area. Are output in parallel. For example, as shown in FIG. 6, five lines of data VA (that is, VA1, V
A2, VA3, VA4, and VA5) are sequentially shifted in output timing using delay elements, and output as data VA11 to VA55 in five columns, respectively (see FIG. 10). Thus, the data VA3 is output as the signals VA31 to VA35 by the four delay elements. Similarly, as shown in FIG. 7, VB of five lines (that is, VB
B1, VB2, VB3, VB4, and VB5) are respectively output as data VB11 to VB55 in five columns using delay elements (see FIG. 11), and as shown in FIG. , VC2, VC3, VC
4, VC5) are output as data VC11 to VC55 in five columns using delay elements (see FIG. 12).
As shown in FIG. 9, five lines of VD (ie, VD
1, VD2, VD3, VD4, and VD5) are output as five columns of data VD11 to VD55 using delay elements (see FIG. 13).

FIG. 14 shows the area determination unit 412 for the small area data VA. (Note that the small area data VB, V
The area determination unit 412 for C and VD is similarly configured. ) For the edge determination, the small area data VA33 of the central pixel output from the small division unit 402 and the small area data VA22, VA23, VA24, VA32,
Differences between VA34, VA42, VA43, and VA44 are respectively obtained by the subtractor 4120, and the obtained differences are compared in the comparator 4121 with the threshold EDGREF17-10 on the white side, respectively. Small area data VA22, VA23, VA24, VA32, VA34,
If the negative logic OR gate 4122 determines that the differences between VA42, VA43, VA44 and the small area data VA33 of the central pixel are all smaller than the threshold value, no signal is output from the OR gate 41210 because it is not an edge.

Similarly, the small area data VA33 of the center pixel output from the small division unit 402 and the small area data VA22, VA23, VA24, VA32, VA of the peripheral pixels around the center pixel are output.
34, VA42, VA43, and VA44 are subtracters 4.
The obtained difference is compared by the comparator 4124 with the black-side threshold value EDGREF27-20, respectively. Small area data VA33 of the center pixel and small area data VA22, VA23, VA24, VA of the peripheral pixels.
32, VA34, VA42, VA43, and VA44, all of which are smaller than the threshold value, the negative logic OR gate 412.
5 is not an edge, the OR circuit 41
No signal is output from 210.

Also, the small area data VA33 of the center pixel,
The small area data of the surrounding pixels and the small area data of the surrounding pixels are compared. For example, a difference between the small area data VA33 of the central pixel and the small area data VA11 of the peripheral pixel and a difference between the small area data VA22 and VA11 of the peripheral pixel are obtained by the subtractor 4126.
Is compared with a second threshold value EDGREF37-30 on the black side. If any difference is greater than the threshold, negative logic A
If the ND gate 4128 determines that the signal is an edge, a signal is output through the OR gates 4129 and 41210. Similarly, the difference between the small area data VA33 of the central pixel and the small area data VA13 of the peripheral pixel and the difference between the small area data VA23 and VA13 of the peripheral pixel are compared with threshold values, respectively. The difference between VA33 and the small area data VA15 of the peripheral pixel and the difference between the small area data VA24 and VA15 of the peripheral pixel are respectively compared with the threshold value, and the small area data VA33 of the central pixel and the small area data VA31 of the peripheral pixel are compared. , And the difference between the small area data VA32 and VA31 of the peripheral pixels are respectively compared with the threshold value. The difference between the small area data VA33 of the central pixel and the small area data VA35 of the peripheral pixel and the small area The difference between the data VA34 and VA35 is compared with the threshold value, respectively, and the small area data VA33 of the central pixel and the small area data VA of the peripheral pixels are compared. 1 and the difference between the small area data VA42 and VA51 of the peripheral pixel are compared with threshold values, respectively, and the difference between the small area data VA33 of the central pixel and the small area data VA53 of the peripheral pixel and the small value of the peripheral pixel are compared. The difference between the area data VA43 and VA53 is compared with a threshold value, respectively, and the difference between the small area data VA33 of the central pixel and the small area data VA55 of the peripheral pixel and the difference between the small area data VA44 and VA55 of the peripheral pixel are respectively Compared to the threshold. If any of the differences is larger than the threshold value, it is an edge and a signal is output.

FIG. 15 shows an edge smoothing processing section 40 for performing edge smoothing on the small area data VA.
6 is shown. The edge smoothing processing unit 412 for the small area data VB, VC, VD has the same configuration. In the edge smoothing processing unit 406, the small area data VA3 of the center pixel output from the small division unit 402
3, and small area data VA22, VA23,
VA24, VA32, VA34, VA42, VA43,
A difference from VA44 is obtained by a subtractor 4060, and the obtained difference is respectively used as a thinning pattern threshold ESR.
The EF 17-10 is compared with the comparator 4061.
The comparison results are respectively sent to the edge attribute determination unit 4064, and it is determined whether or not the comparison corresponds to the edge image pattern (thinning pattern) to be thinned shown on the left side of FIG. If the edge smoothing determination condition is satisfied, ESOUTP = 1 is output.

Similarly, the small area data VA33 of the central pixel output from the small division section 402 and the small area data VA22, VA23, VA24, VA32, VA of the peripheral pixels around the central pixel.
34, VA42, VA43, and VA44 are subtracters 4.
062, and the obtained differences are respectively the fattening pattern threshold value ESREF27-20 and the comparator 4
063. Each of the comparison results is sent to the edge attribute determination unit 4064, and it is determined whether or not the pattern corresponds to the pattern of the edge image to be thickened (thickened pattern) shown on the right side of FIG. If the edge smoothing determination condition is met, ESOUTN = 1 is output. The determination results of ESOUTP = 1 and ESOUTN = 1 are output to the selector via the EXOR gate 4065.

On the other hand, the maximum value circuit 4066 and the minimum value circuit 4
067, the data VA and V of the central small area and its surroundings
The maximum and minimum values of B and VC are obtained, and the selector 406
8 is output. The output of selector 4058 is ESOU
Since the selection is made by the TN, the maximum value is output when the pattern is a fat pattern, and otherwise, the minimum value is output. The output value of this selector 4068 is
069. The selector 4069 also receives the data VA33 of the central small area. Therefore, the selector 4069 outputs the maximum value in the case of the fattening pattern, outputs the minimum value in the case of the thinning pattern, and outputs the data of the central small area otherwise. .

FIG. 16 shows conditions for edge smoothing determination in the edge attribute determination section 4064 in the edge smoothing processing section 406. The six conditions on the left are image patterns to be narrowed. Here, E1
E8 represents the comparison result of the peripheral pixels. If any of these six conditions is met, it is determined that the pattern is a thinning pattern, and ESOUTP = 1. The six conditions on the right side are the pattern of the image to be fattened. here,
F1 to F8 represent comparison results of peripheral pixels. If any of these six conditions is satisfied, it is determined that the pattern is a fat pattern and ESOUTN = 1.

FIG. 17 shows an edge smoothing processing section 40.
6 shows an example of the effect of the data processing of No. 6. 600 shown above
When the pixel data of the * 600 dpi pattern is input, as a result of the processing by the edge smoothing processing unit 406, the resolution is increased to the lower 600 * 2400 dpi pattern. This makes the edges smoother,
The edge image becomes more natural.

As described above, in the present embodiment, in the edge smoothing process, the gradation value of the edge portion is
An image is formed at a resolution determined from the maximum value and the minimum value of the gradation values of the peripheral pixels and corresponding to a small area obtained by dividing the pixel to be processed. As specific examples, FIGS. 18 to 21 show examples of changes in the tone value of the pixel of interest due to the edge smoothing process. As shown in FIG. 18, eight pixels (VA22 to VA22) around the target pixel VA33 are used to determine the edge smoothing.
VA44) is used. In this case, data of a small area at a distance of 600 dpi and one dot is used. As shown in FIG. 19, when performing the edge smoothing by the edge smoothing determination, the tone value used to replace the tone value (VA33) of the attention small area is 24
00dpi, a small area (VD23,
VB33) is used. Here, the data VD23
Is data of a small area of a pixel different from the data VA33 and VB33.

Now, a case where edge smoothing is performed on a pattern as shown in FIG. 20 will be described. FIG.
Of the pattern of interest, patterns around the small area of interest are represented by data VA33, VB33, VC33, and VD33.
It can be seen that the pattern is a fat pattern (see FIG. 16). As the gradation value used for replacing the target small region, the maximum value among the gradation values of the adjacent small regions is used. Therefore, VA33, VB33, VC3
3. Of the VDs 33, only the VAs 33 are replaced with gray values from white to black. Therefore, the result of the edge smoothing process is the pattern shown in FIG.

[0026]

According to the present invention, when the edge is expressed with high resolution and smoothly expressed, the gradation value of the small area is determined from the maximum value and the minimum value of the data around the small area. Continuity is guaranteed. Therefore, even in the case of halftone characters and the like, and in the case of characters on a density background, the edge portion is a smooth edge portion having density continuity. Therefore, it is possible to increase the resolution of the edge portion by inputting the multi-valued image data and realize a smooth edge.

[Brief description of the drawings]

FIG. 1 is a cross-sectional view of the overall configuration of a color digital copying machine.

FIG. 2 is a block diagram of an image data control unit.

FIG. 3 is a block diagram of a tone reproduction unit.

FIG. 4 is a block diagram of a line memory block;

FIG. 5 is a block diagram of a part of a small area dividing unit.

FIG. 6 is a block diagram of a part of a small area dividing unit.

FIG. 7 is a block diagram of a part of a small area dividing unit.

FIG. 8 is a block diagram of a part of a small area dividing unit.

FIG. 9 is a block diagram of a part of a small area dividing unit.

FIG. 10 is a diagram showing data output from the circuit of FIG. 6;

FIG. 11 is a diagram showing data output from the circuit of FIG. 7;

FIG. 12 is a diagram showing data output from the circuit of FIG. 8;

FIG. 13 is a diagram showing data output from the circuit of FIG. 9;

FIG. 14 is a diagram of an area determination unit for small area data VA;

FIG. 15 is a block diagram of an edge smoothing processing unit;

FIG. 16 is a diagram showing conditions for edge attribute determination.

FIG. 17 is a diagram showing an example of data processing of an edge smoothing processing unit.

FIG. 18 is a diagram showing pixels used for determining edge smoothing.

FIG. 19 is a diagram showing pixels on which edge smoothing processing is performed.

FIG. 20 is a diagram showing a pattern before edge smoothing processing;

FIG. 21 is a diagram showing a result of edge smoothing processing on the pattern of FIG. 20;

[Explanation of symbols]

402 small area dividing section, 406 edge smoothing processing section, 412 area determining section, 4064 edge attribute determining section, 4069 selector.

 ──────────────────────────────────────────────────続 き Continued on the front page F-term (reference)

Claims (2)

[Claims] A dividing unit that divides each pixel of the input image data into a plurality of small regions; and a dividing unit that divides the input image data into a small region of interest and its surrounding small regions in units of small regions divided by the dividing unit. An attribute determining unit that determines an edge attribute in consideration of data; and a gradation value of the small region in the unit of the small region divided by the dividing unit according to the edge attribute determined by the attribute determining unit. An image processing apparatus comprising: a region and a gradation value setting unit that sets the gradation value from a gradation value of a small region around the region. 2. The tone value setting unit according to claim 1, wherein said tone value of said small area and a maximum value and a minimum value of tone values of said small area and surrounding small areas are determined according to an edge attribute. An image processing device that sets the gradation value of the small area from the middle.