JP2021066155A - Image processing apparatus, image processing method, and program - Google Patents

Abstract

To prevent reproduction from being destabilized based on a halftone density when performing smoothing.SOLUTION: An embodiment of the present invention provides an image processing apparatus comprising acquisition means for acquiring an inclination characteristic of optical scanning, first conversion means for converting image data having a first resolution into image data having a second resolution, shift processing means for performing, for the image data having the second resolution, processing for shifting an image in a second direction at a position in a first direction to be derived depending on the inclination characteristic, second conversion means for converting image data after the shifting by the shift processing means into the image data having the first resolution, and pulse width conversion means for converting, for the image data having the first resolution acquired by the second conversion means, a pixel value of the image data into a value corresponding to an exposure time period of the light source, the first conversion means or the second conversion means performing conversion using a flying geese-shaped region as a unit.SELECTED DRAWING: Figure 8

Description

The present invention relates to a color image forming apparatus having a plurality of color developing means and a means for sequentially transferring a plurality of color images formed by each developing means.

The electrophotographic method is known as an image recording method used in a color image forming apparatus such as a color printer and a color copier. The electrophotographic method is a method in which a latent image is formed on a photosensitive drum using a laser beam and developed with a charged coloring material (hereinafter referred to as toner). The image is recorded by transferring the image of the developed toner to a recording medium and fixing the image.

A color image forming apparatus generally provided with the same number of developing machines and photosensitive drums as the number of toner colors and adopts a tandem method of sequentially transferring images of different colors onto a belt for carrying the images. In a tandem color image forming apparatus, there are a plurality of factors that cause registration deviation between colors.

Factors of registration deviation include lens non-uniformity and mounting position deviation in the deflection scanning device, and mounting position deviation of the polarization scanning device to the color image forming apparatus main body. Due to these factors, the scanning line is tilted or bent, and the degree of bending (hereinafter referred to as a profile) differs for each image forming apparatus (specifically, for each recording engine) and for each toner color. Causes a shift.

As a technique for suppressing such registration deviation, Patent Document 1 measures the inclination and bending size of a scanning line using an optical sensor, corrects the image data so as to cancel them, and then prints the image data. Disclose the method. According to Patent Document 1, electrical correction (so-called digital correction) is performed on the image data, so that mechanical adjustment members and adjustments at the time of assembly are not required. Therefore, the image forming apparatus can be miniaturized, and the registration deviation can be dealt with at low cost.

This electrical registration deviation correction is divided into a correction for each pixel and a correction for less than one pixel. In the correction in units of one pixel, the pixels are shifted in the sub-scanning direction in units of one pixel according to the amount of correction for inclination and bending. When this method is adopted, the bending and inclination are about 100 to 500 μm, and an image forming apparatus having a resolution of 1200 dpi requires an image memory for several tens of lines for correction. In the following description, the position to shift is referred to as a "transfer point".

In the correction of less than one pixel, the gradation value of the image data is adjusted by the pixels before and after the sub-scanning direction. By performing the correction of less than one pixel by such a method, it is possible to eliminate the unnatural step at the transfer point boundary caused by the correction in units of one pixel and to smooth the image.

When the above-mentioned smoothing process is performed on an image that has been subjected to screen processing immediately before printing, pulse width modulation (hereinafter abbreviated as PWM) is performed on the laser beam, and the laser exposure time is set in the sub-scanning direction. Smoothing is performed by gradually switching to. For example, in the case of 0.5 pixel correction of less than 1 pixel, half exposure is performed twice in the vertical direction of the sub-scanning direction.

Japanese Patent Application Laid-Open No. 2004-170755

When smoothing is performed by modulating the laser exposure time using PWM, the edge portion is blurred. Smoothing is performed using this blur. However, due to the influence of this blurring, the small dot reproduction characteristic and the fine line reproduction characteristic are deteriorated. For example, the reproducibility is not stable with small dots or thin lines that make up halftone dots, and the density changes at the transfer point, resulting in uneven appearance.

As a smoothing method, a method of convolving a smoothing filter is generally used, but when this method is adopted, an intermediate signal of light and shade expressing an edge appears around a sharp edge. This intermediate signal is expressed by PWM, but the expression of the intermediate signal by this PWM deteriorates the dot reproduction characteristic and the fine line reproduction characteristic.

Therefore, in view of the above problems, it is an object of the present invention to suppress destabilization of reproduction based on halftone densities when smoothing is performed.

One embodiment of the present invention includes an acquisition means for acquiring the tilt characteristic of optical scanning by a light source, and a first conversion means for converting image data of the first resolution into image data of a second resolution higher than the first resolution. , A shift that performs a process of shifting the image data of the second resolution in units of one pixel in a second direction different from the first direction at a position in the first direction derived according to the tilt characteristic. The processing means, the second conversion means for converting the shifted image data acquired by the shift processing means into the image data of the first resolution, and the first resolution image acquired by the second conversion means. With respect to the data, there is a pulse width modulation means for converting the pixel value of the image data into a value corresponding to the exposure time of the light source, and a pulse width modulation means for correcting the image in units of less than one pixel. However, the first conversion means or the second conversion means is an image processing apparatus characterized in that conversion is performed in units of a geese-shaped region.

According to the present invention, when smoothing is performed, it is possible to suppress destabilization of reproduction based on halftone densities.

The block diagram which shows the structure of the image forming apparatus. Sectional view of image forming apparatus. The figure which shows the profile characteristic in an image forming apparatus. The figure which shows the relationship between the deviation direction in an image forming part, and the correction direction in an image processing part. The figure which shows the data holding method of a profile characteristic. The figure for demonstrating the binarization by the dither method. The figure which shows typically the image of the line extending along the laser scanning direction, and the state of the shift. Flowchart of shift processing. The figure which shows the dither matrix before and after enlargement. The figure which shows typically the image of transfer and interpolation. The figure which shows the sampling shape used at the time of image reduction.

<Image forming unit 101>
Hereinafter, the image forming unit 101 in the color image forming apparatus 100 of the present embodiment will be described. FIG. 1 is a diagram illustrating a configuration of each block related to creating an electrostatic latent image of the electrophotographic color image forming apparatus 100 according to the present embodiment. The color image forming apparatus 100 includes an image forming unit 101 and an image processing unit 102. The image processing unit 102 generates bitmap format image data, and the image forming unit 101 forms an image based on the generated bitmap format image on the recording medium. FIG. 2 is a cross-sectional view of a four-color tandem electrophotographic color image forming apparatus 100 of Cyan, Magenta, Yellow, and Black that employs the intermediate transfer body 28. In addition, in this specification, when the code of each component is distinguished for each color, it is expressed by adding the character which represents each color to the number, and when it is not necessary to distinguish for each color, it is a number. It shall be represented by only. In the following, the four colors of Cyan, Magenta, Yellow, and Black are represented by one letter each of C, M, Y, and K, and the three colors of Red, Green, and Blue are represented by one letter each of R, G, and B. ..

First, the operation of the image forming unit 101 will be described with reference to FIGS. 1 and 2. The image forming unit 101 forms an electrostatic latent image by driving the exposure means according to the exposure time based on the data processed by the image processing unit 102, and develops the formed electrostatic latent image to develop a monochromatic toner image. To form. Then, the image forming unit 101 superimposes the monochromatic toner images to form a multicolor toner image, transfers the formed multicolor toner image to the recording medium 11 of FIG. 2, and multicolors on the recording medium. The toner image is fixed by heating and pressurizing.

As shown in FIG. 2, the image forming unit 101 has four injection chargers 23Y, 23M, 23C, 23K, and four photoconductors 22Y, 22M, 22C, 22K. Further, the injection charger 23Y has a sleeve 23YS, the injection charger 23M has a sleeve 23MS, the injection charger 23C has a sleeve 23CS, and the injection charger 23K has a sleeve 23KS. With such a configuration, in the image forming unit 101, the photoconductor for each color is charged in the order of Y, M, C, and K.

The photoconductors 22Y, 22M, 22C, and 22K rotate by transmitting the driving force of a driving motor (not shown). The drive motor rotates the photoconductors 22Y, 22M, 22C, and 22K in the counterclockwise direction in the drawing according to the image forming operation. The exposure means is to irradiate the photoconductors 22Y, 22M, 22C, and 22K with exposure light from the scanner units 24Y, 24M, 24C, and 24K, and selectively expose the surfaces of the photoconductors 22Y, 22M, 22C, and 22K. It is configured to form an electrostatic latent image.

As a configuration for visualizing the electrostatic latent image formed on the photoconductor, the image forming unit 101 has four developing devices 26Y, 26M, 26C, and 26K. The developer 26Y has a sleeve 26YS, the developer 26M has a sleeve 26MS, the developer 26C has a sleeve 26CS, and the developer 26K has a sleeve 26KS. With such a configuration, the image forming unit 101 develops each of Y, M, C, and K colors. The developing devices 26Y, 26M, 26C, and 26K can be attached to and detached from the image forming unit 101, respectively.

The intermediate transfer body 28 in FIG. 2 rotates clockwise in the figure in order to receive a monochromatic toner image from the photoconductor 22, and the primary transfer rollers 27Y located opposite to the photoconductors 22Y, 22M, 22C, and 22K. A monochromatic toner image is transferred with the rotation of 27M, 27C, and 27K. By applying a bias voltage to the primary transfer roller 27 and making a difference between the rotation speed of the photoconductor 22 and the rotation speed of the intermediate transfer body 28, the monochromatic toner image is efficiently transferred onto the intermediate transfer body 28. This is called "primary transcription".

Further, the monochromatic toner image for each CMYK color is superimposed on the intermediate transfer body 28. The superimposed multicolor toner image is conveyed to the secondary transfer roller 29 as the intermediate transfer body 28 rotates. At the same time, the recording medium 11 is narrowly conveyed from the paper feed tray 21 to the secondary transfer roller 29, and the multicolor toner image on the intermediate transfer body 28 is transferred to the recording medium 11. At this time, the toner image is electrostatically transferred by applying a bias voltage to the secondary transfer roller 29. This is called "secondary transcription". The secondary transfer roller 29 comes into contact with the recording medium 11 at the position 29a while transferring the multicolor toner image onto the recording medium 11, and is separated from the position 29b after the printing process.

When the recording medium 11 is conveyed to the secondary transfer roller 29, a slight skew occurs. The image transferred by this skew may be skewed on the recording medium 11. Normally, this skew is so small that it is hard to notice, but when cutting a double-sided print, it can be recognized as a misalignment on the front and back.

The fixing device 31 has a fixing roller 32 for heating the recording medium 11 for melting and fixing the multicolor toner image transferred to the recording medium 11 to the recording medium 11, and a fixing roller 32 for pressing the recording medium 11 against the fixing roller 32. It has a pressure roller 33 and. The fixing roller 32 and the pressure roller 33 are formed in a hollow shape. The fixing roller 32 has a built-in heater 34, and the pressurizing roller 33 has a built-in heater 35. The fixing device 31 conveys the recording medium 11 holding the multicolor toner image by the fixing roller 32 and the pressure roller 33, and fixes the toner on the recording medium 11 by applying heat and pressure.

The recording medium 11 after the toner is fixed is then discharged to a paper ejection tray (not shown) by an ejection roller (not shown), and the image forming apparatus 100 ends the image forming operation. The cleaning means 30 cleans the toner remaining on the intermediate transfer body 28. After transferring the multicolor toner image formed on the intermediate transfer body 28 to the recording medium 11, the cleaning means 30 collects the waste toner remaining on the intermediate transfer body 28, and the collected waste toner is used as a cleaner container. It is stored in.

<Profile characteristics>
Hereinafter, the profile characteristics of the scanning lines for each color in the image forming apparatus 100 will be described with reference to FIGS. 3 to 5. FIG. 3A shows a region shifted upward with respect to the laser scanning direction as a profile characteristic of the image forming apparatus, and FIG. 3B shows a region shifted upward with respect to the laser scanning direction as a profile characteristic of the image forming apparatus with respect to the laser scanning direction. Indicates a region that is shifted downward. Reference numeral 301 in the drawing is an ideal scanning line, and indicates a characteristic when laser scanning is performed perpendicular to the rotation direction of the photoconductor 22.

In the following description, the image processing unit 102 will describe the direction to be corrected in the forward direction with respect to the profile, but the definition of the profile characteristic is not limited to this. That is, the profile characteristic may be defined as the deviation direction in the image forming unit 101, and the image processing unit 102 may configure the image forming apparatus 100 so as to perform correction in the opposite direction.

FIG. 4 shows the correlation between the direction in which the image processing unit 102 should be corrected and the deviation direction in the image forming unit 101 according to the profile definition. Specifically, when the bending characteristic as shown in FIG. 4A is shown as the direction in which the image processing unit 102 should be corrected, the profile characteristic of the image forming unit 101 is in the opposite direction of the bending characteristic. It becomes as shown in a certain FIG. 4 (b). Alternatively, when the bending characteristic as shown in FIG. 4C is shown as the profile characteristic of the image forming unit 101, the direction to be corrected by the image processing unit 102 is the opposite direction of the bending characteristic. It will be as shown in FIG. 4 (d).

Further, as a method for retaining profile characteristic data, for example, as shown in FIG. 5, the pixel position in the main scanning direction of the transfer point and the direction of change in the amount of deviation in the sub-scanning direction until the next transfer point are determined. A method of storing the table to be held can be mentioned. Specifically, in the case of FIG. 5, transfer points P1, P2, P3, ... Pm are defined for the profile characteristics of FIG. 5 (a). Each transfer point is a point at which a deviation of one pixel occurs in the sub-scanning direction, and the deviation direction may change upward or downward until the next transfer point.

For example, the transfer point P2 is a point at which the transfer should be performed upward until the next transfer point P3. Therefore, the transfer direction at the transfer point P2 is upward (expressed as ↑) as shown in FIG. 5 (b), and similarly, the transfer direction at the transfer point P3 is also upward (expressed as ↑). On the other hand, the transfer direction at the transfer point P4 is a downward direction (expressed as ↓) unlike the conventional upward direction. As a method of holding the data related to such a direction, for example, "1" may be assigned as the value indicating the upward direction and "0" may be assigned as the value indicating the downward direction. In this case, the direction data in FIG. 5 (b) is represented as shown in FIG. 5 (c). Further, the number of bits of the retained data is only the same as the number of transfer points, and when the number of transfer points is m, the number of bits of the retained data is m.

The bending characteristics 302 of FIGS. 3A and 3B show the positional accuracy and diameter deviation of the photoconductor 22, and the optical system in the scanner units 24Y, 24M, 24C, and 24K of each toner color shown in FIG. Shows the actual scan line with tilt and bend due to position accuracy. This profile characteristic is different for each recording device (specifically, the recording engine) of the image forming apparatus, and further, in the case of the color image forming apparatus, the profile characteristic is also different for each toner color.

Next, the transfer point in the region where the laser scanning direction is shifted upward will be described with reference to FIG. 3A. The transfer point in the present embodiment indicates a point at which the bitmap image data needs to be shifted (shifted) by one pixel because a deviation of one pixel is accumulated in the sub-scanning direction. That is, in FIG. 3A, the points P1, P2, and P3 that are shifted by one pixel in the sub-scanning direction on the upward bending characteristic 302 correspond to the transfer points. In FIG. 3A, the point P0 is used as a reference. As can be seen from FIG. 3A, the distances L1 and L2 between the transfer points become shorter in the region where the bending characteristic 302 changes rapidly, but become longer in the region where the bending characteristic 302 changes slowly.

Next, a transfer point in a region shifted downward in the laser scanning direction will be described with reference to FIG. 3 (b). Even in the region showing the characteristic shifted downward, the definition of the transfer point indicates the point shifted by one pixel in the sub-scanning direction as in the region showing the characteristic shifted upward. That is, in FIG. 3B, the points Pn and Pn + 1, which are shifted by one pixel in the sub-scanning direction on the downward bending characteristic 302, correspond to the transfer points. In addition, also in FIG. 3 (b), as in FIG. 3 (a), the distances Ln and Ln + 1 between the transfer points are shortened in the region where the bending characteristic 302 is abruptly changed, but are gradually changed. It gets longer in the area.

As described above, the transfer point is closely related to the degree of change in the bending characteristic 302 of the image forming apparatus. Therefore, in an image forming apparatus having a sharp bending characteristic, the number of transfer points is large, and conversely, in an image forming apparatus having a gentle bending characteristic, the number of transfer points is small. Further, as already described, since the bending characteristic of the image forming apparatus differs for each color, the number and position of transfer points also differ for each color. This difference between colors will appear as a registration shift in the image in which the toner images of all colors are transferred onto the intermediate transfer body 28.

<Processing executed by the image processing unit 102>
Hereinafter, the processing executed by the image processing unit 102 in the image forming apparatus 100 will be described with reference to FIG.

The image generation unit 104 generates raster image data that can be printed based on print data received from a computer device (not shown) or the like, and outputs it as RGB data for each pixel. The image generation unit 104 may be configured to provide reading means inside the image forming device and handle the image data acquired by reading the reading means, instead of the image data included in the print data received from the computer device or the like. good. The reading means referred to here includes a CCD (Chaerged Couple Device) or a CIS (Contact Image sensor). The image forming apparatus may be configured so as to have a processing unit that performs predetermined image processing on the image data read by the reading means. Alternatively, the image forming apparatus may be configured so that the image data is received from the reading means via an interface (not shown) instead of providing the reading means inside the image forming apparatus. This image data is generated at a resolution corresponding to the exposure resolution of the exposure means. For example, if the scanning line spacing is 1200 dpi, it is generated as image data having a resolution of 1200 dpi.

The color conversion processing unit 105 converts the RGB data received from the image generation unit 104 into CMYK data according to the color of the toner used in the image forming unit 102. This CMKY data is stored in a storage unit 106 having a memory (so-called bitmap memory) for storing bitmap image data. The storage unit 106 is a first storage unit configured in the image processing unit 102, and the storage unit 106 temporarily stores raster image data to be printed. The storage unit 106 may be configured as a page memory for storing image data for one page, or may be configured as a band memory for storing data for a plurality of lines.

The HT processing units 107C, 107M, 107Y, and 107K perform half-toning processing on the data of each color output from the storage unit 106 (specifically, for example, an image in which each pixel has a value of 8-bit 256 gradations). .. As a result, the gradation expression of the input is converted into a pseudo halftone (for example, two gradation) expression represented by the dither method. In this specification, the HT processing unit that uses the dither matrix to reduce the pixel value is also referred to as the dither processing unit.

Here, the binarization by the dither method will be described with reference to FIG. The input continuous gradation image (for example, an image in which each pixel has an 8-bit 256 gradation value) is divided into N × M (8 × 8 in the example of FIG. 6) blocks. After that, the gradation value of the pixels in the block is compared with the threshold value in the dither matrix in which N × M threshold values of the same size as this block are arranged for each pixel, the magnitude relationship is determined, and the determination result is obtained. Output based on. For example, if the pixel value is larger than the threshold value, 1 is output, and if the pixel value is equal to or less than the threshold value, 0 is output. By performing such processing for all pixels for each size of the matrix, it is possible to binarize the entire image.

In an electrophotographic color image forming apparatus, a dither matrix in which a plurality of pixels are concentrated is periodically used in order to realize stable dot reproducibility on paper. On the contrary, if the dots are diffused or if there are many isolated dots that do not have dots in the surroundings, stable gradation reproducibility cannot be obtained due to the instability of dot reproducibility. A screen having a small dot size, a narrow dot spacing, and a large number of dots for expressing the same density is called a high-line number screen, and is used to enhance the sense of resolution. On the contrary, in the case of a low line number screen, each dot size is large, the dot spacing is wide, the number of dots for expressing the same density is small, and smooth gradation expression becomes possible.

The image shift unit 108 performs a process of calculating the coordinates to be shifted, an image shift process, and an interpolation process of a step generated at the shift portion as a process for correcting the deviation of the scanning line of the laser. Details of these processes will be described later.

The storage unit 109 is a second storage unit configured in the image processing unit 102, and the storage unit 109 stores the image data after the shift by the image shift unit 108.

The PWM unit 113 that performs pulse width modulation executes a process of converting the image data for each toner color read from the storage unit 109 into image data in which each pixel has a value equivalent to the exposure time of the scanner unit 114 as a pixel value. To. Then, the PWM unit 113 outputs the converted image data to the scanner unit 114 of the image forming unit 101. By such pulse width modulation, the exposure time of one pixel can be controlled, and the above-mentioned intermediate signal can be generated. That is, by allocating half the exposure time of one pixel as the exposure time, it is possible to generate pixels having a density of 50%.

The scanner unit 114 exposes the drum 115 by irradiating the laser beam according to the pulse width-modulated exposure time signal.

The profile characteristic data already described with reference to FIG. 5 is held in the storage unit 103 in the image forming unit 101 as an inherent characteristic of the image forming apparatus 101. As shown in FIG. 1, profiles 116C, 116M, 116Y, and 116K corresponding to each toner color are stored in the storage unit 103 of the image forming apparatus 101. The image processing unit 102 performs image processing according to the profile characteristics held by the image forming unit 101.

<Step interpolation by simple image shifting and smoothing>
Hereinafter, the step interpolation processing by simple image shifting and smoothing executed by the image shifting unit 108, and the problems thereof will be described.

The image shift unit 108 is oriented perpendicular to the laser scanning direction at the coordinate position obtained by the profile characteristics with respect to the image (binarized image) after low gradation by the half toning process in the HT processing unit 107. Shift processing (offset) is performed to correct the geometry of the image.

In this geometric correction, first, the image shift unit 108 derives each coordinate (position in the main scanning direction) of the transfer point and the correction direction for each transfer point as information on the coordinates to be shifted. Since the coordinates to be shifted are independent coordinates for all four CMYK colors, the coordinate information needs to be derived independently for each of the four CMYK colors.

Subsequently, the image shift unit 108 shifts the image by one pixel in the sub-scanning direction according to the derived coordinate information. After that, the image shift unit 108 performs interpolation processing of less than one pixel on the shifted image. In interpolation of less than one pixel, the gradation value for the pixel of interest in the image data is adjusted by using the pixels before and after the pixel of interest in the sub-scanning direction. As a result, it is possible to eliminate an unnatural step at the boundary of the transfer point caused by the image shift in units of one pixel and to smooth the image. However, this smoothing requires pixels at an intermediate level (pixels representing a density between white and black) with respect to pixels represented by binary values. Such intermediate level pixels can be generated by adjusting the light emission time by PWM.

FIG. 7A is an image of a line having a width of 2 pixels parallel to the laser scanning direction, and shows an image shifted in the sub-scanning direction orthogonal to the laser scanning direction at the center thereof. FIG. 7B shows the result of performing interpolation processing by applying PWM to the laser emission time with respect to the image of FIG. 7A. As shown in FIG. 7B, intermediate levels are generated on the left and right sides of the center of the laser scanning direction. Specifically, as a signal for the region 702, a signal is generated in which the light emission time for one pixel is narrowed down to 75% by PWM, and as a signal for the area 703, a signal in which the light emission time for one pixel is narrowed down to 25% by PWM is generated. Occurs. Similarly, as a signal for the area 704, a signal is generated in which the light emission time for one pixel is narrowed down to 75% by PWM, and as a signal for the area 705, a signal in which the light emission time for one pixel is narrowed down to 25% is generated by PWM. .. By combining the images of the regions 702 to 705 and the image of the region 701 formed with a light emission time of 100%, the step is suppressed (interpolated) while maintaining the line width of two pixels.

The graphs shown in FIGS. 7 (c) to 7 (e) are obtained by plotting the PWM pulse emission time with respect to the laser scanning direction on the horizontal axis. The graph of FIG. 7 (c) shows the state of the pulse generated when forming the image of the region 701. The graph of FIG. 7D shows the state of the pulse generated when forming the images of the regions 702 and 704. The graph of FIG. 7 (e) shows the state of the pulse generated when forming the images of the regions 703 and 705.

In FIG. 7 (c), the double-headed arrow 706 corresponds to the light emission time of one pixel, which is 75% of the light emission time in FIG. 7 (d) and 25% of the light emission time in FIG. 7 (e). There is. In the line shown in FIG. 7 (b), the line image by the pulse of FIG. 7 (d) and the line image of FIG. 7 (e) when the line image by the pulse of FIG. 7 (c) is centered with the shift position as a boundary. The positional relationship with the line image due to the pulse of is changed (that is, the top and bottom are switched). By performing such image processing, the total line width can be saved and smooth interpolation processing can be performed.

As a specific content of the above image processing, when the coordinates in the main scanning direction are the coordinates of the region 702 (region 705), the weight for the pixel of interest is 0.75 and the weight for the pixel in the sub scanning direction is 0.25. Calculate the weighted average to do. Further, when the coordinates in the main scanning direction are the coordinates of the region 703 (region 704), the weighted average is calculated with the weight for the pixel of interest being 0.25 and the weight for the pixel in the sub-scanning direction being 0.75. In this way, the weight counts may be exchanged up and down with the coordinates to be shifted as the center.

However, when such image shift and interpolation processing is performed, as shown in FIG. 7B, pixels represented at an intermediate level are always generated up and down in the sub-scanning direction. For lines of sufficient width as illustrated in FIG. 7 (b), the input line width can be expected to be preserved and stably reproduced even if intermediate level pixels are used, but thinner thin lines and isolated lines can be expected. For small dots, stable reproduction is difficult due to blurring using this intermediate level. Therefore, a means for making thin lines and isolated dots more stable is required. It should be noted that the isolated dots are mainly generated in the halftone region, particularly in the extremely highlighted portion by the half toning process, and the reproduction instability of the isolated dots becomes apparent as uneven density.

<Image shift with image enlargement and smoothing with image reduction>
Hereinafter, regarding image shift and smoothing in the present embodiment, specifically, image shift accompanied by image enlargement and smoothing by image reduction, which are performed to solve the above-mentioned problem of reproduction instability of isolated dots. , FIG. 8 will be described.

The HT processing unit 107 performs the enlargement processing in step S801 and the binarization processing in step S802. In the following, "step S-" will be abbreviated as "S-".

More specifically, in S801, the input image is magnified twice in each of the main scanning direction and the sub-scanning direction, and then in S802, the image obtained by the enlargement is binarized. .. For example, when an image of 1200 dpi is input, the image of 2400 dpi is enlarged in S801, and the image of 2400 dpi is binarized in S802 using a dither matrix. In the following, this specific case will be described.

The image enlargement of S801 is a simple padding, and one pixel is increased to a total of four pixels by two pixels vertically and horizontally while keeping the same pixel value. Further, since the dither matrix used in S802 after S801 is applied at the resolution of the image enlarged in S801 (2400 dpi in this example), an area four times the matrix size assumed at 1200 dpi is required. Here, the dither matrix used in S802 will be described with reference to the examples of FIGS. 9 (a) and 9 (b).

The dither matrix shown in FIGS. 9 (a) and 9 (b) is obtained by cutting out the central portion of the dither threshold matrix shown in FIG. 6, and the dither matrix of FIG. 9 (a) remains cut out. It is the one before the enlargement, and the dither matrix of FIG. 9B is the one after the enlargement. In the dither matrix of FIG. 9 (b), each pixel of the dither matrix of FIG. 9 (a) is not simply enlarged to 4 pixels, but has a minimum threshold value in the sub-scanning direction assuming subsequent reduction processing. There is only one location, and the size of the minimum isolated point is changed. When such a dither matrix is used, the length of the binarized dot in the sub-scanning direction is always odd.

After the processing by the HT processing unit 107, the image shifting unit 108 performs the image shifting processing in S803 and the reduction processing in S804. More specifically, in S803, the image geometrically corrected by performing shift processing (offset) at the coordinate position obtained by the profile characteristics described above with respect to the image binarized in S801 to S802.

Although overlapping with the above description, first, as information on the coordinates to be shifted, each coordinate of the transfer point (position in the main scanning direction) and the correction direction for each transfer point are derived. At this time, the resolution of the input image is twice the image resolution used in the above description. Therefore, since the physical size of one pixel is halved, the shift amount when one pixel is shifted is halved, and the number of shifts is double the number of times described above. Since the coordinates to be shifted required in S803 are independent coordinates for all four CMYK colors, the coordinate information needs to be derived independently for each of the four CMYK colors. Then, at the obtained coordinates, the image is shifted in the sub-scanning direction orthogonal to the laser scanning direction. The shift at this time is a shift at the resolution of the image enlarged in S801 (2400 dpi in this example).

After that, in S804, a reduction process of returning to the original (before enlargement) resolution (1200 dpi in this example) is performed. By reducing the resolution (in this example, it is reduced to 1/2 from 2400 dpi to 1200 dpi), it is possible to apply a process equivalent to the interpolation of less than one pixel described above. As a method of reduction, PWM light emission based on the average value of 4 pixels included in the 2 × 2 region, in other words, PWM light emission according to the number of pixels having a pixel value of 1. Specifically, among the 4 pixels of 2 × 2, if the number of pixels (the number of black pixels) with a pixel value of 1 is 1, 25%, if the number of pixels is 2, 50%, the number of pixels is If it is 3, 75%, and if the number of pixels is 4, 100% PWM light emission is performed.

Here, for reference, the result of applying the above-mentioned reduction smoothing process to the midtone region of the polar highlight and the line image is compared with the result of applying the simple image shift and smoothing described above. This will be described with reference to FIG.

10 (c) and 10 (d) show the results of applying the reduction smoothing process according to the present embodiment. Specifically, FIG. 10 (c) shows before reduction, and FIG. 10 (d) shows after reduction (that is, after smoothing). In this example, the reduction is performed in units of the 2 × 2 rectangular area 101 in FIG. 10 (c). 10 (a) and 10 (b) show the results of applying the simple shift and smoothing described above, as opposed to FIGS. 10 (c) and 10 (d). Specifically, FIG. 10 (a) shows before smoothing, and FIG. 10 (b) shows after smoothing.

In both FIGS. 10 (a) and 10 (b), from the top, the minimum isolated dot generated by half-toning the polar highlight, the isolated dot of the next size, and finally extending laterally. A line with a width of 2 dots is drawn. In each of these images, one pixel is shifted in the sub-scanning direction at the center position in the lateral direction. Further, in the image of FIG. 10 (c), since the resolution is doubled with respect to the image of FIG. 10 (a), a shift occurs twice at both ends.

With respect to the isolated dots, when the above-mentioned simple shift and smoothing are applied, as shown in FIG. 10 (b), intermediate level dots are generated and spread in the sub-scanning direction, and the dot concentration is impaired and the reproducibility is impaired. Is not stable. On the other hand, when smoothing by reduction is applied, as shown in FIG. 10D, it is possible to suppress the generation of dots at an intermediate level and suppress the spread of dots. In addition to this, as shown in FIG. 10 (d), the change in the dot shape before and after the shift is suppressed to the change only in the vertical inversion between the 100% pixel (black pixel) and the intermediate level pixel. It is improved compared to 10 (b).

On the other hand, with respect to the line portion having a 2-dot width, when the above-mentioned simple shift and smoothing are applied, the line width changes uniformly before and after the shift, as shown in FIG. 10 (b). On the other hand, when the smoothing by reduction described this time is applied, as shown in FIG. 10D, an area composed of 100% pixels (black pixels) and intermediate level pixels and 100% pixels (black) are applied. The area consisting of only pixels) is nested at the shift coordinates. This result is obtained because the line width is the same as the reduction processing unit (2 pixels in this example), the phase shifts due to the shift, and the interference occurs, and the width is always a multiple of 2 due to the enlargement. It becomes an unavoidable phenomenon in the line part. Even if there is interference, the density is preserved digitally, but it is difficult to maintain the uniformity of the width of those lines with different values and widths after print output. Therefore, the problem of line width non-uniformity remains even in smoothing by image reduction in units of 2 × 2 pixels.

The problem of non-uniform line width that remains even with this smoothing method by reduction is that when calculating the 4-pixel average during image reduction, the sampling shape is changed from a rectangular shape (2 × 2 square in the above example) to a goose-shaped shape (Z type). It can be solved by changing to a shape (also called a hook shape). Hereinafter, the method for solving this problem will be described in detail.

<Image shift with image enlargement and smoothing by image reduction using flying geese sampling>
Hereinafter, the image shift accompanied by image enlargement and the smoothing by image reduction using flying geese paradigm sampling, which are performed to solve the problem of non-uniform line width, will be described again with reference to FIG. To do.

Similar to the above-mentioned reduction (smoothing) by sampling in units of 2 × 2 pixels, the HT processing unit 107 enlarges the input image twice in each of the main scanning direction and the sub scanning direction in S801, and steps. In S802, a binarization process is performed on the enlarged image. It should be noted that this image enlargement is a simple padding as in the above. Further, here as well, the case where the image of 1200 dpi is enlarged to the image of 2400 dpi will be described.

Here, the dither matrix used in S802 will be described with reference to the examples of FIGS. 9 (a) and 9 (c). The dither matrix shown in FIGS. 9 (a) and 9 (c) is obtained by cutting out the central portion of the dither threshold matrix shown in FIG. 6, and FIG. 9 (a) shows the cut-out portion before enlargement. FIG. 9 (c) is an enlarged view of a mode for solving the problem of non-uniform line width. Like the matrix of FIG. 9 (b) described above, the matrix of FIG. 9 (c) is not simply a four-fold enlargement of the dither matrix of FIG. 9 (a). That is, assuming the subsequent reduction processing, specifically, the flying geese-shaped sampling shape used at the time of reduction, the size of the minimum isolated point is changed and the shape of the minimum isolated point is changed to be slanted. When such a dither matrix is used, the sub-scanning direction length of the dots as a result of binarization becomes an odd number, as in the dither matrix of FIG. 9B described above.

After the processing by the HT processing unit 107, the image shifting unit 108 performs the image shifting processing in S803 and the reduction processing in S804. More specifically, in S803, the image is geometrically corrected by performing shift processing (offset) at the coordinate position obtained based on the profile characteristics of the enlarged and binarized image in S801 to S802. ..

Details of S803 are omitted because they overlap with the previous contents, but for the images enlarged and binarized in S801 to S802, the image shift unit 108 first provides information on the coordinates to be shifted as information on the transfer point. The direction of correction for each coordinate and transfer point is derived. Here, as in the above example, a case where the image of 1200 dpi is enlarged to the image of 2400 dpi in S801 will be described. After deriving the coordinate information, the image is shifted in the sub-scanning direction orthogonal to the laser scanning direction at the obtained coordinates.

After that, in S804, as in the above description, the reduction process for returning to the original (before enlargement) resolution (1200 dpi in this example) is performed. By reducing the resolution in this way (in this example, it is reduced to 1/2 from 2400 dpi to 1200 dpi), it is possible to apply a process equivalent to step interpolation of less than one pixel.

The sampling shape for averaging used at the time of this reduction will be described with reference to FIG. In the present embodiment, the region of the gantry-shaped sampling shape as shown in FIG. 11B is not the average value of the four pixels included in the rectangular sampling shape region 111 as shown in FIG. 11A. The average value of 4 pixels included in 112 is used.

Here, for reference, the result of applying the smoothing process by reduction using flying geese paradigm to the midtone region of the polar highlight and the line image is the average of 4 pixels included in the 2 × 2 rectangular region described above. It will be described with reference to FIG. 10 in comparison with the result of applying smoothing by reduction using a value.

10 (e) and 10 (f) show the results of applying the smoothing process by reduction using the flying geese paradigm sampling. Specifically, FIG. 10 (e) shows before reduction, and FIG. 10 (f) shows after reduction (that is, after smoothing). In this example, the reduction is performed in units of the 4-pixel geese-shaped region 102 in FIG. 10 (e).

As shown in FIG. 10 (e), the isolated dots in the highlight portion appearing in the halftone processing are intentionally slanted, and the orientation and size are matched with the flying geese paradigm shape used at the time of reduction to suppress the spread after reduction. Can be done. If the number of black pixels among the four pixels included in the flying geese paradigm region 112 as shown in FIG. 11B is 1, 25% of PWM light emission is performed, and if the number is 2, 50% of PWM light emission is performed. If the number is 3, 75% PWM emission is performed, and if the number is 4, 100% PWM emission is performed.

For isolated dots of the size next to the smallest isolated dot, two pixels are laid out diagonally in the matrix used in the half toning process, and by averaging diagonally at the same angle, 100% PWM emission is performed on one side in the sub-scanning direction. It becomes possible to create a concentration range. Therefore, in the polar highlight portion, the spread of isolated dots is suppressed as in FIG. 10 (d) described above, and the generation of dots at an intermediate level is also suppressed to the same extent as in FIG. 10 (d).

On the other hand, regarding the line portion having a 2-dot width, when the smoothing by reduction using the 4-pixel average of the above-mentioned 2 × 2 rectangular region is applied, there is a problem of non-uniform line width. The problem is solved when smoothing by reduction using a 4-pixel average of the mold region is applied. As shown in FIG. 10 (f), the same result as the result of applying the step interpolation processing by the simple shift and smoothing described above (see FIG. 10 (b)) can be obtained.

As a result, due to such reduction by geese-type sampling, reduction using the average of 2 × 2 pixels is achieved along with the problem of uneven density of halftones due to the instability of isolated dots, which was a problem in simple shift and smoothing. It solves the problem of line width non-uniformity. From FIG. 10 (f), it can be seen that the reproduction of the isolated dots and the line width uniformity are compatible with each other by the reduction by the flying geese paradigm sampling.

As described above, regarding the interpolation of the step by the pixel-by-pixel shift accompanying the geometric correction of the laser scanning, the method of using the image shift accompanied by the image enlargement and the image reduction by the gantry sampling has been described. According to this method, it is possible to achieve both stability of isolated dots and line width uniformity in the halftone region while performing step interpolation to less than one pixel using PWM.

In the above-described embodiment, the case where simple padding is performed at the time of enlargement and averaging is performed by flying geese paradigm sampling at the time of reduction has been described, but after enlarging to the flying geese paradigm at the time of enlargement, the average of 2 × 2 pixels at the time of reduction is described. May be carried out. That is, the same effect can be obtained even if the shapes are exchanged between enlargement and reduction. However, the dither matrix used at that time needs to be a matrix in consideration of sampling in 2 × 2 units described in FIG. 9B.

Further, in the above-described embodiment, the case where the shift processing and the interpolation processing are performed only in the sub-scanning direction has been described, but the directions of these processes are not limited to the sub-scanning direction. That is, the shift process and the interpolation process may be performed in the main scanning direction. Further, although the exposure by laser scanning has been described as an example, the same effect can be expected for an exposure means that uses an LED or the like as a light source and does not perform laser scanning. In this exposure, LEDs are arranged in a row in contact with the photoconductor for the number of pixels on one side of the image, and the rotating photoconductor is exposed. In this case, the deviation for each color caused by the assembly accuracy of the LED row corresponds to the bending of the laser scanning.

[Other Embodiments]
The present invention supplies a program that realizes one or more functions of the above-described embodiment to a system or device via a network or storage medium, and one or more processors in the computer of the system or device reads and executes the program. It is also possible to realize the processing. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions. The present invention may be configured by appropriately combining the elements of the above-described embodiments.

102 ... Image processing unit 107 ... HT processing unit 108 ... Image shift unit 113 ... PWM unit

Claims (13)

An acquisition means for acquiring the tilt characteristics of optical scanning by a light source,
A first conversion means for converting image data of the first resolution into image data of a second resolution higher than the first resolution, and
Shift processing for shifting the image data of the second resolution in units of one pixel in a second direction different from the first direction at a position in the first direction derived according to the tilt characteristic. Means and
A second conversion means that converts the image data after shifting by the shift processing means into the image data of the first resolution, and
It is a pulse width modulation means that converts the pixel value of the image data into a value corresponding to the exposure time of the light source with respect to the image data of the first resolution acquired by the second conversion means. Pulse width modulation means that corrects in units of less than a pixel,
Have,
The first conversion means or the second conversion means is an image processing apparatus characterized in that conversion is performed in units of a flying geese-shaped region. Further, it has a dither processing means for reducing the pixel value of the image data of the second resolution acquired by the first conversion means.
The image processing apparatus according to claim 1, wherein the shift processing means processes the image data after low gradation acquired by the dither processing means. The first conversion means expands one pixel into four pixels in a 2 × 2 rectangular area.
The image processing apparatus according to claim 2, wherein the second conversion means averages the pixel values of the pixels included in the flying geese-shaped region. The first conversion means performs a process of enlarging an image in an oblique direction with the flying geese-shaped region as a unit.
The image processing apparatus according to claim 2, wherein the second conversion means averages the pixel values of the pixels included in the 2 × 2 rectangular region. The dither processing means applies a dither matrix to the image data of the second resolution acquired by the first conversion means.
The dither matrix is characterized in that, of a plurality of threshold values arranged along the second direction at a predetermined position in the first direction, only one of the plurality of threshold values is the smallest. The image processing apparatus according to 3. The image processing apparatus according to claim 5, wherein in the dither matrix, the smallest of a plurality of threshold values arranged along the second direction is diagonally arranged. The image processing apparatus according to claim 6, wherein in the dither matrix, there is a region in which the same threshold value is continuous and has the same shape as the flying geese paradigm shape. The image processing apparatus processes image data for use in an image forming apparatus that forms an image on a recording medium by light scanning from a plurality of the light sources corresponding to a plurality of toner colors.
One of claims 1 to 7, wherein each of the acquisition means, the first conversion means, the shift processing means, and the second conversion means independently performs processing on each of the plurality of light sources. The image processing apparatus according to. The conversion of the image data by the first conversion means is a conversion that doubles the resolutions of the first direction and the second direction, respectively.
The conversion of image data by the second conversion means is a conversion in which the resolutions of the first direction and the second direction are each halved, according to any one of claims 1 to 8. The image processing apparatus described. The first direction is the main scanning direction of the light source.
The image processing apparatus according to any one of claims 1 to 9, wherein the second direction is a sub-scanning direction orthogonal to the first direction. An acquisition means for acquiring the tilt characteristics of optical scanning by a light source,
A shift processing means for shifting an image of first resolution image data in units of one pixel in a second direction different from the first direction at a position in the first direction derived according to the tilt characteristic. When,
A first conversion means for converting image data after shifting by the shift processing means into image data having a second resolution lower than the first resolution, and
It is a pulse width modulation means that converts the pixel value of the image data into a value corresponding to the exposure time of the light source with respect to the image data of the second resolution acquired by the first conversion means. Pulse width modulation means that corrects in units of less than a pixel,
Have,
The first conversion means is an image processing apparatus characterized in that conversion is performed in units of a flying geese-shaped region. Steps to acquire the tilt characteristics of optical scanning by a light source,
The first conversion step of converting the image data of the first resolution into the image data of the second resolution higher than the first resolution, and
Shift processing for shifting the image data of the second resolution in units of one pixel in a second direction different from the first direction at a position in the first direction derived according to the tilt characteristic. Steps and
A second conversion step of converting the image data after the shift by the shift processing step into the image data of the first resolution, and
In the pulse width modulation step of converting the pixel value of the image data into a value corresponding to the exposure time of the light source with respect to the image data of the first resolution acquired by the second conversion step, the image is converted into 1 A pulse width modulation step that corrects in units of less than a pixel,
Have,
An image processing method characterized in that, in the first conversion step or the second conversion step, conversion is performed in units of a flying geese-shaped region. A program for causing a computer to perform the method according to claim 12.

Images