JPH1065907A - Picture density converter - Google Patents

Abstract

PROBLEM TO BE SOLVED: To inexpensively obtain a pixel density converted picture of a high quality picture by performing density conversion of an input binary picture by using a high densifying means, a low densifying means and a thinning means without making a binary picture multi-valued. SOLUTION: First, the pixel density of an input binary picture is highly densified in a double high densifying ratio which is an integer multiple by means of smoothing processing which is performed by a high densifying part 2. Then, the pixel density is converted into the one in a high densifying ratio of an integer multiple (double) × a low densifying ratio (<3> /4 times) =<3> /2 times = 1.5 times by thinning one pixel every three pixels in vertical and horizontal directions separately in a low densifying part 3. Thereby, a multi-valued line buffer becomes unnecessary. Furthermore, when each pixel that constitutes a binary picture which is performed density conversion coincides with a thinning pattern, each pixel is converted from a black pixel (pixel value 1) to a white pixel (pixel value 0) and performed thinning processing. By this, when an input binary picture is a half-tone picture like a pseudo medium gradation picture, etc., the painting out of a picture can be prevented and a density conversion picture of a high quality picture is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image density conversion device for converting the pixel density of an input binary image and outputting the converted image.

[0002]

2. Description of the Related Art In the case where the readable pixel density of a binary image input device such as a scanner (the number of pixels in a unit area) and the output pixel density of a binary image output device such as a plotter or a display are different. For example, when the pixel density at which the binary image input device can read is 200 dpi and the pixel density at which the binary image output device can output is 300 dpi, the original image read by the binary image input device is used as it is. When output by the binary image output device, since the pixel density is inconsistent, the output image is the original image at 200 dp.
The image is reduced to i / 300 dpi = ２.

Therefore, in order for the image output device to output the original image read by the image input device at the same spatial size, the image density must be increased between the image input device and the image output device. With the conversion device interposed, the pixel density becomes 20
It is necessary to convert an original image of 0 dpi to a pixel density of 300 dpi.

As a method of converting the pixel density, there is a method of simply interpolating pixels by repeating the same pixel according to the conversion ratio of the density. For example, when converting an original image of 200 dpi to a pixel density of 300 dpi, the pixel density can be converted at 1.5 times magnification by repeating the same pixel every other pixel.

However, in such a simple conversion method,
The image quality of the image after the pixel density conversion is insufficient, such as jagged oblique lines. Therefore, Japanese Patent Application Laid-Open No.
As in the technique disclosed in Japanese Patent Application Laid-Open Publication No. H10-115, there has been proposed a technique for improving the image quality of an image after pixel density conversion.

[0006]

However, in Japanese Patent Application Laid-Open No. Hei 5-40826, an edge enhancement process is performed on data obtained by enlarging a binary image by an integral multiple and converting binary to multi-level density. The multi-valued image is reduced and scaled, and the result is re-binarized to obtain a pixel density converted image. In order to perform a scaling process on the multi-valued image converted from the binary, a multi-valued image is obtained. A line buffer is required, which increases costs. Further, although the pseudo halftone portion can obtain a good scaled image, there is a problem that the edges of the character portion are dull due to the multi-valued image, and irregularities are likely to be generated when re-binarized.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and has as its object to provide an image density conversion apparatus capable of obtaining a high-quality image after pixel density conversion at low cost.

[0008]

According to a first aspect of the present invention, there is provided an image density conversion apparatus for converting the pixel density of an input binary image and outputting the converted pixel density. High-density means for converting an image into a high-density binary image by densifying the image at a predetermined integer multiple densification rate by smoothing processing, and reducing the density of the high-density binary image at a predetermined low density rate And a low-density means for converting the low-density binary image to a low-density binary image by performing a thinning process on the low-density binary image.

According to a second aspect of the present invention, in the image density converting apparatus according to the first aspect, the low-density means is configured to generate the high-density binary image at a period corresponding to the predetermined low-density ratio. While the pixel forming the pixel is converted to the low-density binary image by performing the thinning process by thinning out the pixels, the thinning unit performs the thinning process on the low-density binary image. Among the original pixels constituting the input binary image, the pixels obtained by increasing the density of each of the original pixels by the high-density means are other than the original pixels subjected to the thinning-out processing in the low-density means. The pixels after the densification by the densification means with respect to the original pixel are subjected to the thinning processing.

According to a third aspect of the present invention, in the image density converting apparatus according to any one of the first and second aspects, the thinning means is smoothed by a smoothing process in the high density means. An exclusion pattern for retaining a step of a diagonal line is provided, and among pixels constituting the low-density binary image, a pixel that matches the exclusion pattern is not subjected to thinning processing.

According to a fourth aspect of the present invention, in the image density converting apparatus according to any one of the first to third aspects, the thinning means includes: Is characterized in that thinning is performed from the edge of.

According to a fifth aspect of the present invention, in the image density converting apparatus according to any one of the first, second, and third aspects, the thinning means includes the one of the low-density binary image.
It is characterized in that thinning is performed only from the edge in the direction.

[0013]

Embodiments of the present invention will be described below in detail with reference to the accompanying drawings.

FIG. 1 is a diagram showing a block configuration of an image density conversion device according to an embodiment of the present invention. The image density conversion device according to the embodiment of the present invention shown in FIG. 1 is applicable to a facsimile device, an image scanner device, a printer device, and the like. The input unit 1 includes a scanner device for reading a document, a facsimile reception device. , And an interface device for a higher-level device such as an image processing device, for inputting binary image data.
In the present embodiment, it is assumed that the pixel density of the binary image data input from the input unit 1 is 200 dpi.

The densification unit 2 increases the pixel density of the binary image input from the input unit 1 by a predetermined integer multiple by a smoothing process by interpolation between pixels (in this embodiment, twice). The density is increased. The low-density unit 3 converts each of the pixels forming the binary image whose density has been increased by the high-density unit 2 into a predetermined low-density ratio (3/4 times in the present embodiment).
The density can be reduced by thinning out at a cycle corresponding to. The thinned portion 4 has a lower density 2 by the lower density portion 3.
A thinning process is performed to reduce the black line width of the value image. The output unit 5 corresponds to an interface device or the like for a host device such as a printer device, a facsimile plotter device, and an image processing device. In the present embodiment,
The pixel density of the binary image output to the output unit 5 is 300d
pi. The control unit 6 is a microcomputer that controls each of the above units.

Here, the pixel density of the binary image input from the input unit 1 is 200 dpi, and the pixel density of the binary image output from the output unit 5 is 300 dpi. It is necessary to perform 1.5 times the pixel density conversion on the binary image.

Hereinafter, specific processing of pixel density conversion in the image density conversion apparatus according to the embodiment of the present invention will be described separately for the first to fifth embodiments.

First, in the first embodiment, the pixel density of an input binary image is increased by an integer multiple of twice as high as an integer multiple by a smoothing process performed by the densification unit 2, and then the density is reduced. In the unit 3, the pixel density is reduced by an integer multiple (2 times) by reducing the density at a 3 / 4-fold reduction rate by thinning out one pixel every three pixels in each of the vertical and horizontal directions. × Density reduction rate (3/4 times) = 3/2 times =
Convert 1.5 times.

Thus, since the density conversion can be performed without changing the binary image to a multi-valued binary image, a multi-valued line buffer is not required. In addition, by using the smoothing process at the density ratio of an integral multiple as the high-density process, the jagged oblique lines are smoothed and the character image quality is improved.

Further, each pixel constituting the binary image whose density has been converted by the high-density section 2 and the low-density section 3 is shown in FIG.
When the pattern matches the thinning pattern shown in (1), the pixel is converted from a black pixel (pixel value 1) to a white pixel (pixel value 0), and thinning processing is performed. That is, when each pixel is set as a target pixel and the target pixel is a black pixel (pixel value 1), and the target pixel is placed at the center of the thinning pattern of FIG. 2A or 2B, FIG. 2 (a) or (b)
In this case, the pixel of interest is converted into a white pixel as a pixel constituting a boundary between the white pixel region and the black pixel region. In each of the thinning patterns shown in FIG. 2, pixels for which no value is written are indefinite pixels, and are pixels that are not referred to in the thinning processing.

Accordingly, when the input binary image is a halftone image such as a pseudo halftone image, the halftone which tends to increase due to the density conversion by the high-density unit 2 and the low-density unit 3 Since the ratio of black pixels in the image can be reduced, the image can be prevented from being crushed, and a high-quality density-converted image can be obtained.

Next, a second embodiment will be described. Second
The embodiment relates to a low-density binary image in which the density is increased at a double-density ratio in the high-density unit 2 and one pixel is thinned out every three pixels in the low-density unit 3. , About the thinning processing in the thinning unit 4.

In other words, when performing the thinning process on the low-density binary image, the thinning unit 4 performs high-density processing among the original pixels constituting the binary image input from the input unit 1. The pixels obtained by the densification of each of the original pixels by the densification unit 2 are used for the densification by the densification unit 2 for the original pixels other than the original pixels subjected to the thinning processing in the low-density reduction unit 3. The subsequent pixels are to be subjected to the thinning processing.

This will be specifically described with reference to FIG. FIG. 2A shows a pixel arrangement of an input image having a pixel density of 200 dpi input from the input unit 1, and FIG.
FIG. 4A shows a pixel arrangement of a high-density image having a pixel density of 400 dpi obtained by densifying the input image of FIG. FIG. 3B shows a pixel arrangement of a low-density image having a pixel density of 300 dpi obtained by reducing the density of the high-density image in FIG. It should be noted that in FIG. 3B or FIG. 3C, the pixel group included in each thick frame (all 4 pixels of 2 × 2 in FIG. 3B, 2 × 2 pixels in FIG. 4 pixels, 2 pixels vertically aligned, 2 pixels horizontally
Pixel, or one pixel), each of the original pixels constituting the input image shown in FIG.
Further, it shows that the data is obtained by the conversion by the density reduction by the density reduction unit 3.

Now, in FIG. 2B, the pixel matrix is represented by f
(X, y). However, (x = 0, 1, 2, ...),
(Y = 0, 1, 2,...), And, for example, the upper left pixel is specified by the pixel matrix f (0, 0).

In this case, the low-density section 3 performs the pixel matrix f (x, x, x) in the high-density image shown in FIG.
In y), a pixel row satisfying the condition of x = 4n + 2 (n = 0, 1, 2,...) is set as a thinning target row, and the condition of y = 4n + 2 (n = 0, 1, 2,...) By thinning out the filled pixel rows as thinning target rows, a low-density image shown in FIG.

Further, in FIG. 3C, the pixel matrix is g (x, y). Where (x = 0, 1, 2,
..) And (y = 0, 1, 2,...), For example, the upper left pixel is specified by a pixel matrix g (0, 0).

In this case, the thinning unit 4 applies the pixel matrix g (x, x, x) in the low-density image shown in FIG.
In y), a pixel row that satisfies the condition of x = 3n (n = 0, 1, 2,...) is set as a thinning target row, and a condition of y = 3n (n = 0, 1, 2,...) And the pixels belonging to both the thinning target column and the thinning target row, that is, only the pixels at the positions where the thinning target column and the thinning target row intersect are thinned. As a pixel to be thinned, thinning processing is performed by pattern matching with the thinning pattern shown in FIG.

Considering the arrangement of the thinning target pixels, the thinning target pixels belong to a 2 × 2 pixel group of 4 pixels in the low-density image shown in FIG. . Other than the four-pixel group, the two-pixel group arranged vertically, the two-pixel group arranged horizontally, or the one-pixel group is composed of the original pixels constituting the input binary image shown in FIG. 2 is obtained by thinning out a pixel group of 2 × 2 pixels obtained by increasing the density by 2 in the column direction and / or the row direction by the low-density reducing unit 3. On the other hand, a pixel group of 4 pixels to which the thinning target pixel belongs is a 2 × 2 pixel obtained by densifying the original pixels constituting the input binary image shown in FIG. Among the two-pixel group, the application of the thinning processing by the low-density reducing unit 3 is avoided.

The thinning processing can reduce the black pixel ratio of the halftone image after the density conversion and can prevent the image from being collapsed. On the other hand, in the density conversion processing of the image, the original input image before the density conversion has the same size. Black pixel is the high density part 2
The non-uniform size (2 × 2 4)
(Pixels, two vertically arranged pixels, two horizontally arranged pixels, or one pixel), the image after density conversion tends to have a grainy feeling.

However, in the second embodiment, in the thinning processing in the thinning section 4, only the pixels included in the pixel group of 2 × 2 4 pixels are set as the thinning target pixels.
The number of 2 × 2 four-pixel black pixels is reduced, the arrangement of the black pixels constituting the image after density conversion becomes uniform, and the sense of roughness of the image can be suppressed.

Next, a third embodiment will be described. In the third embodiment, the thinning unit 4 is provided with an exclusion pattern shown in FIG. 4 for retaining the steps of the oblique lines smoothed by the smoothing process in the high-density unit 2. Each pixel constituting the low-density image is set as a target pixel, and when the peripheral pixels of the target pixel match the above-described exclusion pattern, even if it matches the thin line pattern shown in FIG. It is not a target.

Here, the exclusion pattern shown in FIG. 4 will be described. 2A to 2H satisfy the conditions of each thinning pattern shown in FIG. 2, the target pixel of the black pixel (pixel value 1) at the center of the pattern forms a step of oblique lines. Therefore, if the line is thinned just because it coincides with the thinning pattern shown in FIG. 2, the step of the oblique line is enlarged, and the image quality becomes negative.

Therefore, in the third embodiment, even when each pixel forming the low-density image from the low-density section 3 matches the thinning pattern shown in FIG. When the pixel is coincident with the exclusion pattern, by not thinning the pixel, it is possible to prevent the diagonal steps in the image after the density conversion from being enlarged, and to perform the thinning processing while maintaining the smoothness.

This will be described with reference to FIG. In the figure, (a) shows a signal 2 input from the input unit 1.
The shaded portion of the value image is shown. The shaded portion in (a) is subjected to a smoothing process by the densification unit 2 and the density is increased at a double densification rate, as shown in (b). In (b), a row to be thinned out in the low-density reducing unit 3 and
The rows to be thinned in the thinning unit 4 in the light of the thinning pattern shown in FIG.
The oblique lines shown in (c) are obtained by thinning out the thinning target rows shown in (b) by the low density unit 3. Here, if each pixel constituting the thinning target row is thinned just because it matches the thinning pattern shown in FIG.
As shown in (d), the step of the oblique line becomes large.

However, in the third embodiment, even if each pixel constituting the thinning target row matches the thinning pattern shown in FIG. 2, but matches the exclusion pattern shown in FIG. Therefore, the smoothness of the oblique lines is maintained as shown in FIG.

Next, a fourth embodiment will be described. In the fourth embodiment, the pixel density of the binary image input from the input unit 1 is facsimile DTL (7.7 lines / mm × 8 pixels / mm) or STD (3.85 lines / mm × 8 pixels / m).
m), the output device of the output unit 5 is constituted by a printer with a pixel density of 300 dpi, and the thinning unit 4 performs thinning from edges in two orthogonal directions of the low-density image from the low-density unit 3. It is. For example, the thinning processing is performed as shown in FIG.
2B is performed by pattern matching using a thin line pattern for cutting the upper edge and a thin line pattern for cutting the left edge in FIG. 2B.

Thus, the input binary image from the input unit 1 is a facsimile image, and the output device is a pixel density of 300 dpi.
In the case of the electrophotographic printer described above, since the thinning processing is performed from the edges in two directions, it is optimal for suppressing crushing and roughness, and the best image can be obtained.

Next, a fifth embodiment will be described. In the fifth embodiment, the pixel density of the binary image input from the input unit 1 is facsimile DTL (7.7 lines / mm × 8 pixels / mm) or STD (3.85 lines / mm × 8 pixels / m).
m), the output device of the output unit 5 is constituted by a printer having a pixel density of 360 dpi, and the thinning unit 4 performs thinning only from edges in one direction of the low-density image from the low-density unit 3. is there. For example, the thinning processing is performed by pattern matching using only a thinning pattern for cutting the upper edge in FIG.

Accordingly, the input binary image from the input unit 1 is a facsimile image, and the output device is a pixel density of 360 dpi.
In the case of the ink jet printer described above, the thinning processing is performed only from the edge in one direction, so that it is optimal for suppressing the crushing and roughness, and the best image can be obtained.

In the above-described embodiment, the case where the densification rate in the high-density section 2 is twice and the density reduction rate in the low-density section is 3/4 times has been described. But,
The present invention is not limited to this, and can be similarly applied to a combination of an arbitrary density multiple of an integral multiple and an arbitrary density reduction.

[0042]

According to the first aspect of the present invention, the input binary image is converted into a high-density binary image by a high-density ratio of a predetermined integral multiple by a smoothing process by a high-density means. Then, the high-density binary image is converted into a low-density binary image by reducing the density at a predetermined density reduction rate by the low-density means, and the thinning processing is performed on the low-density binary image by the thinning means. Is performed, the density conversion can be performed in a binary state without converting the binary image into a multi-valued image, and a multi-valued line buffer becomes unnecessary. In addition, by using the smoothing process at the density ratio of an integral multiple as the high-density process, the jagged oblique lines are smoothed and the character image quality is improved. In addition, the thinning process can reduce the ratio of black pixels in a halftone image, which tends to increase due to density conversion, thereby preventing image collapse. Therefore, a high-quality image after pixel density conversion can be obtained at low cost.

According to the second aspect of the present invention, the low-density reducing means thins out pixels forming the high-density binary image at a cycle corresponding to the predetermined low-density rate, thereby reducing the low-density image. When the thinning processing is performed on the low-density binary image, the thinning unit converts the original pixels constituting the input binary image. The pixels obtained by increasing the density of each of the original pixels by the high-density means are used by the high-density means for the original pixels other than the original pixels subjected to the thinning processing in the low-density means. Since the pixels after the density increase are to be subjected to the thinning processing, the black pixels of the same size in the original input image before the density conversion are formed by the high-density means and the low-density means. Density reduction by thinning It is possible to suppress the propensity to become non-uniform size, arrangement of black pixels constituting the image after the density conversion is uniform, it is possible to suppress the roughness of the image.

According to the third aspect of the present invention, the thinning means includes an exclusion pattern for retaining a step of a diagonal line smoothed by the smoothing process in the high-density means, Of the pixels that make up the image, those that match the exclusion pattern are not subject to thinning processing, so the black pixels that make up the diagonal steps are converted to white pixels by thinning, and the diagonal steps are converted. Makes it possible to optimize the ratio of black pixels by the thinning while preventing the adverse effect of the thinning, which degrades the image quality of the image after the density conversion.

According to the fourth aspect of the present invention, the thinning means thins the low-density binary image from edges in two directions orthogonal to each other. When a printer having a pixel density of 300 dpi is used, by performing thinning processing from edges in two directions,
It is optimal for suppressing crushing and roughness, and the best image can be obtained.

According to the fifth aspect of the present invention, the thinning means performs thinning only from edges in one direction of the low-density binary image, so that the input binary image is converted to a facsimile image,
When the output device is a printer with a pixel density of 360 dpi, the thinning processing is performed only from the edge in one direction,
It is optimal for suppressing crushing and roughness, and the best image can be obtained.

[Brief description of the drawings]

FIG. 1 is a diagram showing a block configuration of an image density conversion device according to an embodiment of the present invention.

FIG. 2 is a diagram showing a thinning pattern referred to in a thinning process in the image density conversion device according to the embodiment of the present invention.

FIG. 3 is a diagram illustrating a transition of a pixel arrangement due to a higher density, a lower density, and a thinner line of an input binary image in the image density conversion device according to the embodiment of the present invention.

FIG. 4 is a diagram showing an exclusion pattern referred to in a thinning process in the image density conversion device according to the embodiment of the present invention.

FIG. 5 is a diagram for explaining a case where image density conversion processing in the image density conversion device according to the embodiment of the present invention is applied to a diagonal image.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Input part 2 High-density part 3 Low-density part 4 Thinning part 5 Output part 6 Control part

Claims (5)

[Claims] 1. An image density conversion apparatus for converting the pixel density of an input binary image and outputting the converted image, wherein the density of the input binary image is increased by a predetermined integer multiple by a smoothing process. A high-density means for converting the high-density binary image into a low-density binary image; a low-density means for converting the high-density binary image into a low-density binary image by reducing the density at a predetermined low-density reduction rate; An image density conversion device, comprising: a thinning unit that performs a thinning process and outputs the result. 2. The image processing apparatus according to claim 1, wherein the low-density means performs low-density processing by thinning out pixels constituting the high-density binary image at a cycle corresponding to the predetermined low-density rate.
While converting to a value image, the thinning means,
In the case where the thinning processing is performed on the low-density binary image, among the original pixels constituting the input binary image, pixels obtained by increasing the density of each of the original pixels by the high-density means are used. 2. The thinning processing is performed on pixels after the high-density processing performed on the original pixels other than the original pixels subjected to the thinning processing by the low-density processing. An image density conversion device as described in the above. 3. The pixel forming the low-density binary image, wherein the thinning unit includes an exclusion pattern for holding a step of a diagonal line smoothed by the smoothing process in the high-density unit. 3. The image density conversion device according to claim 1, wherein a pattern that matches the exclusion pattern is not subjected to thinning processing. 4. The image density conversion according to claim 1, wherein the thinning means performs thinning from edges in two orthogonal directions of the low-density binary image. apparatus. 5. The image density conversion apparatus according to claim 1, wherein said thinning means performs thinning only from edges in one direction of said low-density binary image. .