JP5297347B2 - Pixel number conversion method, program for executing the same, and pixel number conversion apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To shorten a time required for conversion processing to attain acceleration while suppressing the deterioration of image quality. <P>SOLUTION: This pixel number converter 1 is provided with: a pixel type discrimination part 2 which detects change point pixels (contour point pixels or boundary point pixels) included in a binary image (original image 10); a mask processing part 3 which sets a pixel number converted image 20 in which the number of pixels in the main scanning and vertical scanning directions of the binary image is increased or decreased according to conversion magnification, and sets pixels corresponding to the change point pixels on the original image 10 in a mask area 32 to generate a converted mask image 30 out of the pixel number converted image 20; a first gradation acquisition part 4 which calculates a density value of the pixels belonging to the mask area 32 of the converted mask image 30 by density calculation processing using a projection method, and binarizes the density value by binarization processing using an error diffusion method to obtain a gradation value; and a second gradation acquisition part 5 which sets a gradation value of pixels belonging to a non-mask area 33 of the converted mask image 30 to a gradation value which the pixel on the original image 10 corresponding to that pixel has. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to a pixel number conversion method for converting the number of pixels of an input binary image in accordance with a designated conversion magnification, and is expressed in particular using a plurality of halftone dots such as a facsimile image and a newspaper image. The present invention relates to a method suitable for converting the number of pixels of a halftone image.

  In recent years, with the diversification of media devices and the improvement of functions of printing equipment such as CTP and rotary press, there is an increasing need to enlarge / reduce halftone images such as text images and newspaper images. When converting the resolution of an image, an interpolation process is generally required, and this is the same even in the case of a binary image expressed by two gradations of black pixels or white pixels. Interpolation methods include a logical sum method, nearest neighbor method, nine-division method, projection method, inverse distance method, and the like (for example, refer to Non-Patent Document 1). It is known that it is easy to suppress deterioration.

  Here, the basic principle of the interpolation processing by the projection method will be described with reference to FIG. FIG. 16 exemplifies a case where resolution conversion processing in the reduction direction (conversion processing for reducing the number of pixels) is performed on the original image 50. In the figure, pixels of a converted image (hereinafter referred to as “converted image” as appropriate) 60 are drawn with thick lines.

  In the projection method, the converted image 60 is projected (superimposed) on the original image 50, an average density value of the original image 50 in one pixel (target pixel P ′) region of the converted image 60 is obtained, and this is calculated as the target pixel P ′. Is replaced with the density value IP ′. The density value IP ′ can be obtained using the arithmetic expression shown in FIG.

  Specifically, areas S0 to S3 of areas where the target pixel P ′ of the converted image 60 and the pixels P0 to P3 of the original image 50 overlap (for example, the area S0 is the target pixel P ′ of the converted image 60 and the original image). Corresponding to the area of the area where 50 pixels P0 overlap), and in each of the pixels P0 to P3, the areas S0 to S3 are multiplied by the density values IP0 to IP3 of the pixels P0 to P3. Then, the multiplication values at the respective pixels P0 to P3 are added to obtain a sum, and the sum is divided by the sum of the areas S0 to S3 (the area of the target pixel P ′ of the converted image 60) to obtain the average density value IP ′. Ask.

  In the case of the resolution conversion processing of the halftone image, the obtained average density value IP ′ is binarized and the target pixel is distributed to either a black pixel or a white pixel.

  As an example of resolution conversion processing using such a projection method, for example, Patent Document 1 discloses a first reduction unit that performs reduction conversion (reduction) on a pixel data string of an original image based on the projection method, and a first reduction method. A second reduction means for further reducing the pixel data string reduced by the means, and a pixel data string reduced by the second reduction means by density preservation binarization processing (binarization processing using an error diffusion method). A first binarizing means for binarizing; a detecting means for detecting a preset pattern based on the original image; a third reducing means for reducing a data string indicating a detection result obtained by the detecting means; Based on the data string indicating the detection result reduced by the third reduction unit, the pixel value similar to the pixel value when binarizing by the first binarization unit is binarized by simple binarization processing When the pattern is detected by the second binarizing means and the detecting means An image processing apparatus comprising: a selection unit that selects a binarization result obtained by the binarization unit 2 and selects a binarization result obtained by the first binarization unit when a pattern is not detected. Have been described.

  Japanese Patent Application Laid-Open No. 2004-228688 discloses a line buffer that stores image data of an original image in the main scanning direction in units of one line, and the image data stored in the line buffer is weighted in accordance with an area coefficient, and the target pixel A main scanning direction density calculation unit for calculating the density of the image, a temporary buffer for cumulatively storing the pixel density of interest of each line calculated by the main scanning direction density calculation unit, and a plurality of lines stored in the temporary buffer. A resolution conversion apparatus including a sub-scanning direction density calculation unit that adds the target pixel density of the target pixel and the target pixel density of the next line, and a binarization unit that generates a binary image based on the sum of the target pixel density Have been described.

  Further, as an interpolation process using a method other than the projection method, Non-Patent Document 2 focuses on the periodicity of image data and uses Fourier transform, inverse Fourier transform, etc. used in signal processing to obtain image data. It is shown that the resolution conversion process is performed by converting the frequency period.

  Specifically, a binary image is converted into a multi-value image and a discrete Fourier transform process is performed, and surrounding components are cut out and a value is normalized from the Fourier image obtained by the discrete Fourier transform. Thereafter, inverse discrete Fourier transform processing is performed, and the obtained multi-valued image data is binarized to create a binary image after resolution conversion. For binarization processing, fixed binarization, error diffusion, dithering, or the like is used.

Japanese Patent No. 2851724 JP 2001-157040 A

Hiroshi Masashima and 4 others, "Performance evaluation and processing speed improvement methods for various enlargement / reduction methods of binary images", IPSJ Transactions Sep.1985 Vol.26 No.5 Joji Tajima, “Gravitational rearrangement of halftone dots for moire-free color proofing”, Journal of the Institute of Image Electronics Engineers, Vol. 13, No. 5 (2005)

  However, in the conversion methods described in Patent Documents 1 and 2, when converting from an original image to a converted image, density calculation processing using a projection method is executed for all pixels included in one page of image data. There is a need to.

  On the other hand, in the projection method, when obtaining the density value of one pixel of interest P ′, in the case shown in FIG. 16, the coordinates of four pixels P0 to P3 that overlap with the pixel of interest P ′ are obtained. Since calculation processing such as obtaining the areas S0 to S3 of two overlapping regions is required, the amount of calculation tends to increase. In particular, the amount of computation at this time increases as the number of original image pixels overlapping the pixel of interest P ′ increases, so depending on the reduction ratio, a huge amount of computation processing is required and the processing speed is slow. There is.

  Further, the conversion methods described in Patent Documents 1 and 2 are intended for resolution conversion processing at a reduction magnification, and resolution conversion processing at an enlargement magnification and separate reduction magnifications for each of the main scanning and sub-scanning directions. There is a disadvantage that cannot be dealt with when an enlargement magnification is assigned.

  Furthermore, since the resolution conversion process described in Non-Patent Document 2 is a processing method based on information such as period value conversion and black pixel ratio, resolution conversion is performed while maintaining black pixel density and pixel position. There is a problem that can not be. In addition, since this processing method uses two-dimensional discrete Fourier transform and inverse two-dimensional discrete Fourier transform, there is a problem that the amount of calculation of image processing becomes large and the processing speed becomes slow.

  Therefore, the present invention has been made in view of the above-described problems in the prior art, and it is possible to reduce the time required for conversion processing while suppressing deterioration in image quality and to increase the number of pixels. The purpose is to provide a method and the like.

  In order to achieve the above object, the present invention provides a pixel number conversion method for converting the number of pixels of an input binary image in accordance with a designated conversion magnification, and a change point pixel included in the binary image is converted. A first step of detecting and setting a pixel number converted image in which the number of pixels in the main scanning and sub-scanning directions of the binary image is increased or decreased according to the conversion magnification, and among the pixels of the pixel number converted image, A second step of generating a conversion mask image by setting a pixel corresponding to the change point pixel on the binary image as a mask area, and using an interpolation operation for the density value of the pixel belonging to the mask area of the conversion mask image A third step of obtaining a gradation value by binarizing the obtained density value and a gradation value of a pixel belonging to a non-mask area of the conversion mask image corresponding to the pixel. The pixels on the binary image Characterized in that it comprises a fourth step of setting a gradation value.

  According to the present invention, for the pixel corresponding to the change point pixel having a large influence on the image quality of the converted image, the gradation value is obtained through the density calculation process using the interpolation operation, and the converted image is obtained. For pixels corresponding to non-change point pixels that have little effect on image quality, the tone value is obtained by a simple process that takes over the tone value on the input binary image, while suppressing deterioration in image quality, It becomes possible to speed up the conversion process.

  In the pixel number conversion method, in the second step, a pixel on the binary image corresponding to the target pixel on the pixel number conversion image is obtained, and whether or not the pixel corresponds to the change point pixel. Accordingly, it can be determined whether or not to set the target pixel as a mask region. According to this, it is possible to appropriately create a conversion mask image in any conversion process in the reduction direction and the enlargement direction.

  In the pixel number conversion method, the coordinates of the pixel on the binary image corresponding to the pixel of interest can be obtained using the coordinates of the pixel of interest on the pixel number conversion image and the conversion magnification.

  In the pixel number conversion method, the density calculation process using the interpolation calculation can be a density calculation process using a projection method.

  In the pixel number conversion method, in the third step, the density value can be binarized by binarization processing using an error diffusion method. In this case, since the binarization process using the error diffusion method is performed only for the pixel corresponding to the change point pixel, the region where the error is dispersed may be limited to the range of the change point pixel region. It is possible to prevent noise from occurring in the non-change point pixel region due to the influence of error diffusion.

  In the above-described pixel number conversion method, the change point pixel is located at a peripheral portion of a halftone dot and forms a contour point pixel constituting the outline of the halftone dot, or a boundary between the halftone dot and a portion other than the halftone dot It is possible to use a boundary point pixel.

  According to another aspect of the present invention, there is provided a program for executing a pixel number conversion process for converting the number of pixels of an input binary image in accordance with a designated conversion magnification, and the pixel number conversion method according to any one of the above It is for performing.

  Furthermore, the present invention is a pixel number conversion device for converting the number of pixels of an input binary image in accordance with a designated conversion magnification, and detecting a pixel type for detecting a change point pixel included in the binary image. A pixel number conversion image in which the number of pixels in the main scanning and sub-scanning directions of the binary image is increased or decreased according to the conversion magnification, and among the pixels of the pixel number conversion image, A mask processing unit that sets a pixel corresponding to the change point pixel as a mask area to generate a conversion mask image, and obtains a density value of a pixel belonging to the mask area of the conversion mask image by a density calculation process using interpolation. In addition, a first gradation acquisition unit that binarizes the obtained density value to obtain a gradation value, and a gradation value of a pixel that belongs to a non-mask area of the conversion mask image corresponds to the 2 Set to the gradation value of the pixel on the value image Characterized in that it comprises a second gradation acquisition unit that.

  As described above, according to the present invention, it is possible to increase the speed by reducing the time required for the conversion process while suppressing the deterioration of the image quality.

1 is a block diagram showing an embodiment of a pixel number conversion apparatus according to the present invention. It is explanatory drawing of a contour point pixel and a boundary point pixel. It is a figure which shows an example of an original image, a contour point pixel area | region, and a boundary point pixel area | region. It is a figure which shows an example of an original image and a conversion mask image. It is explanatory drawing of a mask process. It is a figure which shows the pixel arrangement | sequence of an original image and a conversion image, Comprising: It is an arrangement | sequence figure in the case of performing the conversion process of an expansion direction. It is a figure which shows the relationship between the pixel of a pixel number conversion image, and the pixel of an original image. It is a figure which shows the relationship between the pixel of a pixel number conversion image, and the pixel of an original image. It is a figure which shows the relationship between the pixel of a pixel number conversion image, and the pixel of an original image. It is a figure which shows the relationship between the pixel of a pixel number conversion image, and the pixel of an original image. It is a figure which shows the pixel arrangement | sequence of an original image and a conversion image, Comprising: It is an arrangement | sequence figure in the case of performing the conversion process of a reduction direction. It is a figure which shows the relationship between the pixel of a pixel number conversion image, and the pixel of an original image. It is a figure which shows the relationship between the pixel of a pixel number conversion image, and the pixel of an original image. It is a flowchart which shows the preparation procedure of a conversion mask image. It is a flowchart which shows the procedure of the acquisition process of a gradation value. It is explanatory drawing of the basic principle of the interpolation process by a projection method.

  Next, modes for carrying out the invention will be described in detail with reference to the drawings.

  FIG. 1 is a block diagram showing an embodiment of a pixel number conversion apparatus according to the present invention. This pixel number conversion apparatus (hereinafter referred to as “conversion apparatus”) 1 is included in an input binary image. A pixel type discriminating unit 2 for detecting change point pixels, a mask processing unit 3 for generating a conversion mask image by masking pixels corresponding to the change point pixels, and gradation values of pixels belonging to the mask area of the conversion mask image A first gradation acquisition unit 4 to be acquired, a second gradation acquisition unit 5 to acquire gradation values of pixels belonging to the non-mask area of the conversion mask image, and a first and second gradation acquisition unit 4 5 and a synthesis processing unit 6 that obtains converted image data for one page by synthesizing the outputs of 5.

  The various processing units 2 to 6 are not necessarily configured by hardware, and a part or all of the processing units 2 to 6 may be configured by software (program).

  A memory 7 in the drawing is a storage medium for storing image data to be processed by the conversion apparatus 1 and storing processing values and image data processed by the processing units 2 to 6. In the memory 7, as image data to be processed by the conversion apparatus 1, the data value of each pixel is a binary gradation value (for example, gradation value = 1 for a black pixel, gradation value = for a white pixel = Binary image data represented by 0) is stored.

  Next, operation | movement of the converter 1 which has the said structure is demonstrated, referring FIGS. Here, individual operations of the pixel type determination unit 2, the mask processing unit 3, the first gradation acquisition unit 4, and the second gradation acquisition unit 5 will be described, and then the overall operation of the conversion apparatus 1 will be described. Will be explained.

  First, the operation of the pixel type determination unit 2 will be described with reference to FIGS.

  The pixel type discriminating unit 2 detects change point pixels included in a binary image for one page (hereinafter referred to as “original image” as appropriate), and classifies the pixels in the original image into change point pixels and non-change point pixels. To do. Here, the change point pixel includes a contour point pixel (black pixel) that constitutes the contour of a halftone dot, and a boundary point pixel (white pixel) that constitutes a boundary between the halftone dot and the base paper (a portion other than the halftone dot). There is.

  When detecting the contour point pixel, as shown in FIG. 2A, an arbitrary black pixel in the original image is set as a target pixel. For example, a range including the target pixel and eight pixels adjacent to the target pixel is set. refer. If at least one of the 8 pixels located around the pixel of interest has a white pixel, the pixel of interest is detected as a contour point. When detecting the boundary point pixel, as shown in FIG. 2B, if any white pixel in the original image is the target pixel and there is at least one black pixel in the adjacent eight pixels, the attention is paid. A pixel is detected as a boundary point pixel.

  In detecting the change point pixel, the number of pixels to be referred to is not limited to 8 pixels, and even if the number of pixels is other than 8 pixels, such as referring to 4 pixels adjacent to the target pixel in the vertical and horizontal directions. Good.

By these processes, as shown in FIG. 3, with respect to dot on the original image 10 HD (FIG. 3 (a)), the contour point pixel region 11 along the periphery HD _E halftone dot HD is detected At the same time (FIG. 3B), the boundary pixel area 12 is detected so as to surround the halftone dot HD (FIG. 3C).

  On the other hand, the non-change point pixel is detected using the detection results of the contour point pixel region 11 and the boundary point pixel region 12 described above, and the internal point pixel region 13 (see FIG. 3) located inside the contour point pixel region 11. (B)) and pixels belonging to the external point pixel region 14 (see FIG. 3C) located outside the boundary point pixel region 12 are detected as non-change point pixels.

  Next, the operation of the mask processing unit 3 will be described with reference to FIGS.

  As shown in FIG. 4, the mask processing unit 3 performs mask processing on the detected change point pixel region 15 including the contour point pixel region 11 and the boundary point pixel region 12 to create a conversion mask image 30.

  In creating the conversion mask image 30, first, as shown in FIG. 4B, based on the specified conversion magnification, the pixel number conversion image 20 having the number of pixels corresponding to the conversion magnification is set. For example, if the number of pixels of the original image 10 is 16 × 16 pixels and the designated conversion magnification is 10/16 (reduction conversion), a 10 × 10 pixel number conversion image 20 is set.

  Next, an arbitrary pixel on the pixel number conversion image 20 is selected as the target pixel P ′, and the coordinates of the pixel on the original image 10 corresponding to the target pixel P ′ are obtained by the following expressions 1 and 2.

  Here, “x” and “y” in Expressions 1 and 2 are the coordinates of the pixel of the original image 10, and “x ′” and “y ′” are the coordinates of the target pixel P ′ of the pixel number conversion image 20. It is. “L” and “L ′” are the number of pixels in the main scanning direction of the original image 10 and the number of pixels in the main scanning direction of the pixel number conversion image 20, respectively. “H” and “H ′” are These are the number of pixels in the sub-scanning direction of the original image 10 and the number of pixels in the sub-scanning direction of the pixel number conversion image 20, respectively. L ′ / L corresponds to the conversion magnification in the main scanning direction, and H ′ / H corresponds to the conversion magnification in the sub-scanning direction.

  Next, as shown in FIG. 5A, a pixel 21 on the original image 10 corresponding to the target pixel P ′ of the pixel number conversion image 20 is referred to, and the pixel 21 is detected as a change point pixel. It is determined whether or not there is. As a result, if the pixel corresponds to the change point pixel, the target pixel P ′ is masked, and conversely, if it corresponds to the non-change point pixel instead of being detected as the change point pixel, Do not mask P '.

  Thereafter, the above-described processing is also performed on the other pixels in the pixel number converted image 20 to determine whether or not to sequentially mask all the pixels on the pixel number converted image 20. Thereby, as shown in FIG.4 (c), the conversion mask image 30 in which the mask area | region 32 was set on the basis of the pixel count after conversion is produced.

  In the case of the conversion process in the enlargement direction, similarly to the above, the pixel 21 on the original image 10 corresponding to the target pixel P ′ on the pixel number conversion image 20 is obtained, and whether the pixel 21 corresponds to the change point pixel. Whether or not the target pixel P ′ is set in the mask area 32 is determined according to whether or not (see FIG. 5B).

  As described above, in the above mask processing, since the corresponding pixel of the original image 10 is obtained from the pixels of the pixel number conversion image 20, the pixel number conversion image can be obtained regardless of the conversion processing in the reduction direction or the enlargement direction. It is possible to determine whether or not to set the mask area 32 for all the pixels on 20, and the conversion mask image 30 can be appropriately created.

  Next, the operation of the first gradation acquisition unit 4 will be described with reference to FIGS. 4 and 6 to 12.

  The first gradation acquisition unit 4 obtains a density value by performing density calculation processing using a projection method on a mask region 32 (see FIG. 4C) on the conversion mask image 30 as a target. The density value is binarized by an error diffusion method to obtain a gradation value.

  In the first gradation obtaining unit 4, one of the pixels belonging to the mask area 32 of the conversion mask image 30 is set as the target pixel P ′, and the density value IP ′ of the target pixel P ′ is set as follows. Ask for.

  Here, first, an example in which the original image 10 having 3 × 3 pixels is converted into an image having 5 × 5 pixels (specified conversion magnification: 5 ÷ 3 = 1.666...) A method of calculating density values when performing number conversion will be described.

  In calculating the density value, as in the case of the projection method shown in FIG. 16, the conversion mask image 30 is projected onto the original image 10, and any one pixel belonging to the mask area 32 on the conversion mask image 30 is the target pixel. Set as P '. Then, an average density value of the original image 10 in the region of the target pixel P ′ is obtained and set as the density value IP ′ of the target pixel P ′.

  FIG. 6 is a schematic diagram showing a pixel arrangement of an original image of 3 × 3 pixels and a converted image of 5 × 5 pixels. As can be seen from FIG. 6, in the case of converting the number of pixels in the enlargement direction, the number of pixels Therefore, the area of one pixel of the conversion mask image 30 is smaller than the area of one pixel of the original image 10.

  Therefore, as a state generated by the relationship between the target pixel P ′ of the conversion mask image 30 and the pixel of the original image 10, for example, depending on the position of the target pixel P ′ in the projected original image 10, (a) the target pixel P In a state where the entire 'is included in the region of one pixel of the original image 10 (see FIG. 7), (b) the pixel of interest P' is included in the region of one pixel of the original image 10 in the sub-scanning direction. On the other hand, in the main scanning direction, the target pixel P ′ intersects two adjacent pixel regions of the original image 10 (see FIG. 8), and (c) the target pixel P ′ is the original image 10 in the main scanning direction. While included in the region for one pixel, in the sub-scanning direction, the state in which the pixel of interest P ′ intersects the region of two adjacent pixels of the original image 10 (see FIG. 9), (d) the pixel of interest P ′ is Four types of states (see FIG. 10) that intersect with an area of 4 pixels (2 × 2 pixels) of the original image 10 are assumed.Hereinafter, a method for calculating the density value IP ′ of the target pixel P ′ will be described for each state.

(A). When the entire pixel of interest P ′ is included in the area of one pixel of the original image 10 (FIG. 7)
In this case, the density calculation process using the projection method is not performed, and the gradation value of the pixel P (x, y) of the original image 10 including the target pixel P ′ (x ′, y ′) is referred to. Note that (x ′, y ′) is a coordinate on the conversion mask image 30, and (x, y) is a coordinate on the original image 10.

  Then, the gradation value of the pixel P (x, y) is set to the gradation value of the target pixel P ′ (x ′, y ′) as it is. Therefore, if the corresponding pixel P (x, y) on the original image 10 is a black pixel (gradation value = 1), the gradation value of the pixel of interest P ′ (x ′, y ′) is set to 1 and black. Conversely, if the pixel P (x, y) on the original image 10 is a white pixel (gradation value = 0), the gradation value of the pixel of interest P ′ (x ′, y ′) is set. Set to 0 for white pixels.

(B). In the sub-scanning direction, the pixel of interest P ′ is included in the region of one pixel of the original image 10, while the pixel of interest P ′ intersects the region of two adjacent pixels of the original image 10 in the main scanning direction. (Fig. 8)
The area of the region where the target pixel P ′ (x ′, y ′) and the pixel P (x, y) of the original image 10 overlap is defined as S (0, 0), and the target pixel P ′ (x ′, y ′) and the original pixel 10 When the area of the region where the pixel P (x + 1, y) of the image 10 overlaps is S (1,0) and the density value of the pixel P (x + 1, y) is IP (x + 1, y), In this case, the density value IP ′ (x ′, y ′) of the target pixel P ′ (x ′, y ′) is obtained by Expression 3. The right side of Equation 3 is a fraction with S (0,0) + S (1,0) as the denominator, but S (0,0) + S (1,0) = 1 (1 of the target pixel P ′ Since it is for pixels), the description is omitted.

(C). In the main scanning direction, the target pixel P ′ is included in the area of one pixel of the original image 10, while the target pixel P ′ intersects with the adjacent two pixels of the original image 10 in the sub-scanning direction. (Fig. 9)
In this case, the density value IP ′ (x ′, y ′) of the target pixel P ′ (x ′, y ′) is obtained by the following Equation 4. Here, S (0,1) in Expression 4 is an area of a region where the target pixel P ′ (x ′, y ′) and the pixel P (x, y + 1) of the original image 10 overlap, and IP ( x, y + 1) is a density value of the pixel P (x, y + 1). Further, since the denominator (S (0,0) + S (0,1)) on the right side is 1 (one pixel of the target pixel P ′) as in the case of Expression 3, the description is omitted. Hereinafter, the same applies to other expressions.

(D). When the target pixel P ′ intersects with an area of 4 pixels (2 × 2 pixels) of the original image 10 (FIG. 10)
In this case, the density value IP ′ (x ′, y ′) of the target pixel P ′ (x ′, y ′) is obtained by the following Equation 5. Here, S (1,1) in Expression 5 is an area of a region where the pixel of interest P ′ (x ′, y ′) and the pixel P (x + 1, y + 1) of the original image 10 overlap, IP (x + 1, y + 1) is a density value of the pixel P (x + 1, y + 1).

  Next, taking the original image 10 with 5 × 5 pixels as an image with 3 × 3 pixels (specified conversion magnification: 3 ÷ 5 = 0.6) as an example, the conversion of the number of pixels in the reduction direction is performed. A method for deriving the density value at the time of performing will be described.

  FIG. 11 is a schematic diagram showing a pixel arrangement of an original image of 5 × 5 pixels and a converted image of 3 × 3 pixels. As can be seen from FIG. 11, when the pixel number conversion in the reduction direction is performed, the number of pixels Therefore, the area of one pixel of the conversion mask image 30 is larger than the area of one pixel of the original image 10.

  Therefore, as a state generated by the relationship between the target pixel P ′ of the conversion mask image 30 and the pixel of the original image 10, for example, depending on the position of the target pixel P ′ in the projected original image 10, (e) the target pixel P A state where “′ intersects with an area corresponding to 4 pixels (2 × 2 pixels) of the original image 10 (see FIG. 12), and (f) a target pixel P ′ is an area corresponding to 9 pixels (3 × 3 pixels) of the original image 10 Two types of intersecting states (see FIG. 13) are assumed. Hereinafter, a method for deriving the density value IP ′ of the target pixel P ′ will be described for each state.

(E). When the target pixel P ′ intersects with an area of 4 pixels (2 × 2 pixels) of the original image 10 (FIG. 12)
In this case, since the relationship between the pixel of interest P ′ and the pixels of the original image 10 is the same as in the case of (d) in the conversion of the number of pixels in the enlargement direction, the density value of the pixel of interest P ′ (x ′, y ′) IP ′ (x ′, y ′) is obtained by Equation 5 above.

(F). When the target pixel P ′ intersects with an area of 9 pixels (3 × 3 pixels) of the original image 10 (FIG. 13)
In this case, the density value IP ′ (x ′, y ′) of the target pixel P ′ (x ′, y ′) is obtained by the following Equation 6. Here, S (2,0), S (2,1), S (0,2), S (1,2), and S (2,2) in Equation 6 are respectively the pixel P of the original image 10. (x + 2, y), P (x + 2, y + 1), P (x, y + 2), P (x + 1, y + 2), P (x + 2, y + 2) and Is the area of the overlapping region. IP (x + 2, y), IP (x + 2, y + 1), IP (x, y + 2), IP (x + 1, y + 2), IP (x + 2, y + 2) are pixels P (x + 2, y), P (x + 2, y + 1), P (x, y + 2), P (x + 1, y + 2), P (x + 2 , y + 2).

  After obtaining the density value of the pixel of interest P ′ (x ′, y ′) as described above (multivalue having a numerical value after the decimal point), the pixel of interest P is binarized using the error diffusion method. A gradation value (value of 1 or 0) of '(x', y ') is obtained, and is distributed to a black pixel (gradation value = 1) and a white pixel (gradation value = 0).

  Next, the operation of the second gradation acquisition unit 5 will be described with reference to FIG.

  The second gradation acquisition unit 5 performs a simple magnification conversion process on an area 33 (non-mask area: see FIG. 4C) other than the mask area 32 on the conversion mask image 30.

  In the simple magnification conversion process, first, one of the pixels belonging to the non-mask area 33 on the conversion mask image 30 is set as the target pixel P ′, and the target pixel P ′ is set by the following equations 7 and 8. The coordinates of the corresponding pixel on the original image 10 are obtained.

  Here, “i” and “j” in Expressions 7 and 8 are the coordinates of the pixel of the original image 10, and “i ′” and “j ′” are the coordinates of the target pixel P ′ of the conversion mask image 30. is there. “L” and “L ′” are the number of pixels in the main scanning direction of the original image 10 and the number of pixels in the main scanning direction of the conversion mask image 30, respectively. “H” and “H ′” are respectively The number of pixels in the sub-scanning direction of the original image 10 and the number of pixels in the sub-scanning direction of the conversion mask image 30.

  Next, the gradation value of the pixel P (i, j) on the original image 10 corresponding to the target pixel P ′ (i ′, j ′) of the conversion mask image 30 is referred to, and the pixel P (i, j) The gradation value is set to the gradation value of the target pixel P ′ (i ′, j ′) as it is. Therefore, if the pixel P (i, j) on the corresponding original image 10 is a black pixel (gradation value = 1), the gradation value of the pixel of interest P ′ (i ′, j ′) is set to 1 and black. Conversely, if the pixel P (i, j) on the original image 10 is a white pixel (gradation value = 0), the gradation value of the target pixel P ′ (i ′, j ′) is set. Set to 0 for white pixels.

  Next, the overall operation of the conversion apparatus 1 will be described with reference to FIGS.

  First, the pixel type determination unit 2 detects change point pixels included in a binary image (original image 10) for one page, and determines change point pixels and non-change point pixels. Next, the mask processing unit 3 sets the pixel number conversion image 20 according to the designated conversion magnification, and determines an arbitrary pixel in the pixel number conversion image 20 as a target pixel.

  Next, as shown in FIG. 14, the coordinates of the pixel of the original image 10 corresponding to the target pixel on the pixel number conversion image 20 are obtained using the above-described formulas 1 and 2 (step S1), and then the obtained coordinates are obtained. It is determined whether or not this pixel corresponds to a change point pixel (contour point pixel or boundary point pixel) (steps S2 and S3). As a result, if the corresponding pixel of the original image 10 corresponds to the change point pixel, the target pixel on the pixel number conversion image 20 is set in the mask area 32 (see FIG. 4) (step S4), and the change point pixel corresponds. Otherwise, the target pixel is set in the non-mask area 33.

  Thereafter, the same processing is performed for the remaining pixels on the pixel number conversion image 20 (step 5), and a conversion mask image 30 (see FIG. 4) is created.

  When the creation of the conversion mask image 30 is completed, as shown in FIG. 15, the conversion mask image 30 is raster scanned (step S11), and it is determined whether or not the scanned pixel (target pixel) is a pixel belonging to the mask area 32. It discriminate | determines (step S12).

  As a result, when the target pixel is a pixel belonging to the mask region 32 (step S12: Y), the first gradation acquisition unit 4 obtains a density value by density calculation processing using a projection method. At this time, if the conversion process is in the enlargement direction for increasing the number of pixels, the density value is obtained by the calculation of Expressions 3 to 5 (step 14). Conversely, if the conversion process is in the reduction direction for reducing the number of pixels. Then, the density value is obtained by the calculation of Expression 5 and Expression 6 (Step S15).

  Next, the density value is binarized by binarization processing using an error diffusion method to obtain a gradation value (step S16), and the processing result is written in the memory 7 (step S17).

  On the other hand, when the target pixel is a pixel belonging to the non-mask area 33 (step S12: N), the second gradation acquisition unit 5 uses the expressions 7 and 8 to obtain the original image corresponding to the target pixel. The coordinates of the pixel on 10 are obtained. Then, referring to the obtained gradation value of the pixel, it is set as the gradation value of the target pixel (step S18), and the processing result is written in the memory 7 (step S17).

  Thereafter, the processing of steps S11 to S18 is executed for the remaining pixels on the conversion mask image 30 (step S19), and the gradation values of all the pixels are acquired.

  Finally, the synthesis processing unit 6 synthesizes the processing result of the first gradation acquisition unit 4 and the processing result of the second gradation acquisition unit 5 to complete the entire converted image.

  As described above, according to the present embodiment, in obtaining the gradation value of the pixel of the converted image, the pixel corresponding to the non-change point pixel that has little influence on the image quality of the converted image is obtained from the original image 10. Since the gradation value is obtained by a simple process that takes over the gradation value described above, it is possible to reduce the time required for the conversion process without going through a calculation process with a large amount of calculation.

  On the other hand, for pixels corresponding to change point pixels that have a large influence on the image quality of the converted image, a gradation value is obtained through a density calculation process using a projection method. The occurrence of white pixel crushing, moire, etc.) can be suppressed.

  As described above, in the present embodiment, the pixels of the original image 10 are classified into change point pixels and non-change point pixels, and the gradation value acquisition method is used according to the characteristics of each pixel type. It is possible to increase the speed of the conversion process while suppressing deterioration.

  Further, according to the present embodiment, when the conversion mask image 30 is created, the pixel 21 on the original image 10 corresponding to the pixel of interest P ′ on the pixel number conversion image 20 is obtained, and the pixel 21 becomes the change point pixel. In order to determine whether or not the pixel of interest P ′ is set in the mask area 32 according to whether or not this is the case, the conversion mask image 30 is appropriately displayed in any of the conversion processes in the reduction direction and the enlargement direction. Can be created.

  Furthermore, according to the present embodiment, since binarization processing using the error diffusion method is performed only for the pixels corresponding to the change point pixels, noise is generated in the non-change point pixel region due to error diffusion. Can be prevented. In other words, since error diffusion is to disperse errors to a plurality of pixels located around the pixel of interest, the application range of binarization processing using the error diffusion method is limited to pixels corresponding to change point pixels. In this case, the region where the error is dispersed can be limited to the change point pixel region 15 (the peripheral portion of the halftone dot) and the vicinity thereof. For this reason, under the influence of the error caused by the binarization process, the non-change point pixel region (the central portion of the halftone dot and the black pixel group or the base paper portion and the white pixel group It is possible to prevent the gradation value of the pixels belonging to the (region) from being inadvertently inverted and causing noise.

  In the above embodiment, the calculation process using the projection method is exemplified as the density calculation process executed by the first gradation acquisition unit 4. However, the density value is obtained by the density calculation process using another interpolation calculation. May be calculated.

  In the above embodiment, after obtaining the gradation value of the pixel corresponding to the change point pixel, the gradation value of the pixel corresponding to the non-change point pixel is obtained. However, these processes are performed in the reverse order. Or may be executed in parallel.

  Further, in the above embodiment, the case where the halftone dot HD is composed of the black pixel group and the base paper portion is composed of the white pixel group is illustrated, but the present invention is not limited to the white paper on the base paper of the black pixel group. The present invention can also be applied to a halftone image in which a halftone dot HD composed of a pixel group is arranged.

DESCRIPTION OF SYMBOLS 1 Pixel number conversion apparatus 2 Pixel type discrimination | determination part 3 Mask processing part 4 1st gradation acquisition part 5 2nd gradation acquisition part 6 Composition processing part 7 Memory 10 Original image 11 Contour point pixel area 12 Boundary point pixel area 13 Internal point pixel area 14 External point pixel area 15 Change point pixel area 20 Pixel number conversion image 21 Original image pixel 32 corresponding to target pixel Mask area 33 Non-mask area
HD dot
HD _E
P Pixel on the original image
P 'pixel of interest
S Area of the area where the target pixel and the original image pixel overlap

Claims (8)

A pixel number conversion method for converting the number of pixels of an input binary image according to a designated conversion magnification,
A first step of detecting change point pixels included in the binary image;
A pixel number conversion image in which the number of pixels in the main scanning and sub-scanning directions of the binary image is increased or decreased according to the conversion magnification is set, and a change point pixel on the binary image among the pixels of the pixel number conversion image A second step of generating a conversion mask image by setting pixels corresponding to the mask region;
A third step of obtaining density values of pixels belonging to the mask area of the converted mask image by density calculation processing using interpolation, and binarizing the obtained density values to obtain gradation values;
And a fourth step of setting a gradation value of a pixel belonging to the non-mask area of the conversion mask image to a gradation value of a pixel on the binary image corresponding to the pixel. Number conversion method.   In the second step, a pixel on the binary image corresponding to the pixel of interest on the pixel number conversion image is obtained, and the pixel of interest is masked according to whether the pixel corresponds to the change point pixel. 2. The pixel number conversion method according to claim 1, wherein it is determined whether or not to set the area.   3. The pixel number conversion according to claim 2, wherein coordinates of a pixel on the binary image corresponding to the target pixel are obtained using the coordinates of the target pixel on the pixel number conversion image and the conversion magnification. Method.   The pixel number conversion method according to claim 1, wherein the density calculation process using the interpolation calculation is a density calculation process using a projection method.   5. The pixel number conversion method according to claim 1, wherein, in the third step, the density value is binarized by binarization processing using an error diffusion method.   The change point pixel is a contour point pixel which is located at a peripheral portion of a halftone dot and constitutes the outline of the halftone dot, or a boundary point pixel which constitutes a boundary between the halftone dot and a portion other than the halftone dot The pixel number conversion method according to claim 1, wherein: A program for executing a pixel number conversion process for converting the number of pixels of an input binary image according to a designated conversion magnification,
The program for performing the pixel number conversion method in any one of Claims 1 thru | or 6. A pixel number conversion device that converts the number of pixels of an input binary image according to a designated conversion magnification,
A pixel type determination unit that detects change point pixels included in the binary image;
A pixel number conversion image in which the number of pixels in the main scanning and sub-scanning directions of the binary image is increased or decreased according to the conversion magnification is set, and a change point pixel on the binary image among the pixels of the pixel number conversion image A mask processing unit that generates a conversion mask image by setting pixels corresponding to the mask region;
A first gradation acquisition unit that obtains a density value of a pixel belonging to the mask area of the conversion mask image by density calculation processing using interpolation and binarizes the obtained density value to obtain a gradation value; ,
A second gradation acquisition unit that sets a gradation value of a pixel belonging to the non-mask area of the conversion mask image to a gradation value of a pixel on the binary image corresponding to the pixel; A pixel number conversion device.

Images