JPS63220670A - Inter-line interprolating method for image signal line - Google Patents

Abstract

PURPOSE:To eliminate that a curved line and a slanting line are expressed in step-forme by executing a slanting line smoothing processing when the color of the picture element of both lines of a certain line direction position is in a special relation, in two adjoining lines of an image signal. CONSTITUTION:The memory contents (picture element string data D(n) of an (n)-th line) of a first line memory 11 are inputted from the line memory 11 to a first register 21, and also the memory contents (picture element string data D(n+1) of an (n+1)-th line) of a second line memory 12 are inputted respectively from the second line memory 12 to a second shift register 22, than both contents are given to a picture element conversion ROM 3, and the data of corresponding interpolating picture element string or the data for writing two times are outputted to a third shift register 23 in the form of the parallel data. The data are outputted to an output terminal OUT as serial data. By repeating the above-mentioned operation, the picture element string data to interpolate the section between two adjoining lines are obtained successively.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a method for creating pixel signals that interpolate between lines (scanning lines) of image signals.

[Prior art]

When a facsimile machine reads a document image, the amount of information is reduced by making the scanning density in the direction perpendicular to the line (sub-scanning density) coarser than the scanning density in the line (scanning line) direction (main-scanning density). A method has been adopted to reduce the

However, when an image signal read in this manner is reproduced, blank spaces naturally occur between lines, causing a so-called sagging phenomenon and deteriorating image quality.

In order to avoid this sagging phenomenon, a technique called so-called double writing, in which the same line is reproduced twice in succession, is generally adopted.

However, with this method, although there is no problem with straight lines in the sub-scanning direction, thin straight lines in the main scanning direction may be expressed as thick, and curves, diagonal lines, etc. may lose their smoothness and be expressed in a step-like manner. However, the overall image is reproduced as unnatural.

[Problem that the invention seeks to solve]

Considering these circumstances, for example,
Inventions such as No. 53 have been proposed. This invention interpolates the blank space between adjacent lines with a line created by the logical product or logical sum of the two lower lines.

However, the method of this invention does not solve the problem that curves and diagonal lines are expressed in a step-like manner.

The present invention was made in view of these circumstances, and
To provide an inter-line interpolation method for image signals in which curves and diagonal lines are not expressed stepwise and are smoothly expressed.

[Failure to solve the problem]

In the present invention, in two adjacent lines of an image signal, pixels in both lines have the same color (same signal value) at a certain line direction position, and the next two or three pixels change in the other line. 1! By performing the diagonal line smoothing process, when the color changes to 3 or 4 pixels ahead of J", and the other line changes to a different color up to 3 or 4 pixels ahead,
This is to create an image signal of a line that interpolates between both lines.

The inter-line interpolation method for image signals of the present invention is performed when reproducing an image signal formed by arranging a plurality of image signal lines in a direction orthogonal to the line direction, in which a plurality of lines of an image signal are formed by arranging a plurality of binary pixel signals in one direction. The binary pixel signals of two adjacent lines are compared sequentially between pixels whose line direction positions correspond, and from the pixel next to the first line direction position where the pixels of both lines both have one signal value, one of the values is When each pixel of a line successively takes the other signal value, and there are 2 or 3 pixels between the pixels of the other line and the second line direction position where the pixel of the other line changes to the other signal value, Regarding the line on the above-mentioned - side, the first
The present invention is characterized in that by converting the next pixel in the line direction position to the other signal value, a pixel signal of a line interpolated between both lines is created by performing diagonal line smoothing processing.

[Effect]

In the present invention, in two adjacent lines of an image signal, the pixels of both lines have the same color (same signal value) at a certain line direction position, and the next two or three pixels do not change in one line. If the color of 1 or 2 pixels on one line (signal value ) is transformed and smoothed.

[Principle of the invention]

First, the principle of the present invention will be explained below.

FIG. 1 is a schematic diagram showing the principle of diagonal line smoothing processing in the interline interpolation method for image signals according to the present invention.

In the present invention, pixel column data D (n
) and D(r+-+1), pixel column data D(n+
'Create A).

Hereinafter, explanation will be given according to the example shown in FIG. 1. First, a pixel serving as a reference for smoothing processing is fixed as follows.

Pixel a. : pixel column data D(n+1
．． ) at the reference or origin change pixel.

At the beginning of each line, the line Immediately before the first pixel of (left side in the diagram) Assume a white pixel located at .

Pixel a1: pixel column data D(n+1) of the n+lth line
The first changed pixel on the upper side of pixel ao in the line direction, that is, in the main scanning direction (hereinafter referred to as the right side in Figure 2).

Pixel b1: pixel column data D inl of the nth line, located on the right side of pixel ao, and pixel a. The first changing pixel that changes to a different color (signal value).

After fixing the pixel a O+ a l+ bl as described in "2" below, the diagonal line smoothing process is performed according to the following rules (1) and (2) only in the case of 1 < Ialbl l<=3. However, la+b+1 is the distance between both pixels a1 and bl, specifically the number of pixels interposed between them.

(1) If pixel S is located on the right side of pixel al, convert the signal value of the next pixel after pixel b1, that is, if the pixel is white, it becomes black, and if it is black, it becomes white. Convert each to

(2) If pixel b1 is located to the left of pixel a1, convert the signal value of pixel b1.

The above will be explained based on the example of FIG. 1 +a+ and according to FIG. 2 fal. In each figure, Z represents a black pixel, and the mouth represents a white pixel.

First, the pixel column data D (n+1
) is assumed to be a white pixel ao (indicated by a broken line in the figure). In this case, pixel a1 is the first black pixel of the pixel column data D (n+1) of the nth line]-1, so it becomes the leftmost pixel, and pixel b1 is the pixel column data D (n+1) of the nth line. It becomes the first pixel of nl.

Therefore, the distance 1a1b1) between both pixels a1 and b1 is "
0'', so the diagonal line smoothing process is not performed.

Next, the same process is performed after advancing one pixel in the scanning direction, but the process in this case will be explained according to FIG.
Since the first black pixel of the pixel string data D(n+1) of the n+'lth line is a(1, pixel a1 is the first white pixel of the pixel string data D(n+1) of the n+lth line, that is, the fourth pixel. The pixel b1 becomes the second pixel (white) of D fnl of the pixel column data of the nth line.

Therefore, the distance la+b,1 between both pixels a1 and b1 is "2
'', the above-mentioned process (11 or (2) is performed. In this case, pixel b1 is located on the left side of pixel al, so the process of (2) is performed, that is, the pixel column data of the nth line is D (invert the signal value (color) of pixel b1 of nl,
From white to black. The pixel column after this process is the first
Figure (interpolated pixel column D (n+%) shown in the center of al
become. Then, both pixel column data [1(n+1) and D
(Newly created interpolated pixel row D(n+%
), a pattern as shown on the right side of ta+ in FIG. 1 is obtained.

By performing similar processing, interpolated pixel rows as shown in the center of each are created from the pixel rows of the two adjacent lines shown on the left side of FIG. 1, fbl to fhl. The interpolated pattern shown on the right side of is obtained.

〔Example〕

Examples of the method of the present invention based on the above-mentioned principle will be described below.

FIG. 3 is a block diagram showing an example of a circuit configuration for carrying out the method of the present invention, which is specifically incorporated into a facsimile machine.

In the figure, IS is a first switching circuit. The first switching circuit S1 is controlled by the clock control circuit 10 to switch the image signal received from another facsimile device or the like via the input terminal IN to the first line memory 11 or the second line memo 1. Switch the output to Month 2.

Both the first line memory 11 and the second line memory 12 have a capacity to store the data of one line of pixel rows of the image signal, and the switching processing of the first switching circuit S1 described above allows the first line memory 11 and the second line memory 12 to Pixel column data D (nl and D(n+1)) of lines n and r1+1 are stored.

The outputs of both line memories 11.12 are connected to the first and second shift registers 21.22, respectively, and are output as output signals from the output terminal OUT via the second switching circuit S2 and the third switching circuit S3. It is also output.

Both shift registers 21 and 22 each convert one line of pixel column data as serial data provided from the line memories 11 and 12 into parallel data and provide the parallel data to the pixel conversion ROM 3.

The pixel conversion ROM 3 outputs an interpolated pixel string as shown in FIG. 1 to the third shift register 23 based on the input pixel string data of both the n-th and (n+1) lines.

Specifically, the pixel conversion ROM 3 has a first memory that has been created in advance according to the principle explained in the above-mentioned [Principle of the Invention] section.
The original pixel array pattern shown in the figure is set as an address input, and data of an interpolated pixel array corresponding to each address input is stored as memory contents. Therefore, the pixel column data of both lines inputted from both shifts 1 and 21 and 22 are as shown in FIG.
(If the state is as shown on the left side of hl, interpolated pixel string data as shown on the center of the figure is output to the third shift l-L-sysk 23.

In addition, if the pixel string data of both lines input from both shift registers 21 and 22 are in the state shown on the left side of FIG. (in this case, the pixel column data I of the nth line) (nl is output as is to the third shift register 23, and the pixel column data D fnl is written twice.

Note that the third interpolated pixel string data from the pixel conversion ROM 3
output to the shift register 23 and the second and third switching circuits S2. The switching control of S3 is performed by the clock control circuit 10.

The operation of such a circuit is as follows.

First, the clock control circuit 10 selects the first switching circuit S1.
is switched and controlled, and pixel column data D (nl) for one line of the n-th line is applied to the first line memory 11 and stored.

Next, the clock control circuit]0 causes the second and third switching circuits S2. S3 is switched and the first line memo IJ1
2 (pixel column data Dine of the nth line) is outputted from the first line memory 11 from the output terminal OUT via the second switching circuit S2 and the third switching circuit S3.

Next, the first switching circuit Sl is switched by the clock control circuit 10, and the pixel column data D(n+1
) is input to the second line memory 12 and written ↑a.

Then, the second switching circuit S is controlled by the clock control circuit IO.
2 is switched, the storage contents of the first line memory 11 (pixel column data I]n of the nth line) are inputted from the first line memory 1] to the first shift register 21, and the 2 line memory 12 (n+2th
Pixel column data [1(n+1)) of the line is input from the second line memory 12 to the second shift 1 register 22, respectively.

The data input to both shift registers 21 and 22 is given to the pixel conversion ROM 3 as a parallel data address, and the data of the corresponding interpolation pixel column or the data for double writing is sent to the third shift register in the form of parallel data. 23. The third shift register 23 outputs the parallel data of the interpolated pixel array given from the pixel conversion ROM 3 to the output terminal 0flT as serial data.

After this, switching circuit S2. By switching S3, the stored contents of the second line memory 12 are output to the output terminal OUT, and then by switching the switching circuit S1, the pixel column data D (n+2) of the n+2 line is input to the first image memory j1. It is written as a. Then, as above, the n+2th
The pixel column data for interpolation between the line and the 04-2nd line, or the data for double writing, is stored in the pixel conversion ROM3.
is output from.

By repeating the above operations, pixel string data for interpolating between two adjacent lines can be sequentially obtained.

FIG. 4 is a schematic diagram showing an example of image data processed by such a circuit. That is, when two lines of pixel column data, n-th and n-+-1, such as in FIG. 4 (al) are input to the input terminal I11, conventionally,
As shown in Figure I), the n-th line was repeatedly written twice, but in the present invention, lines are interpolated by smoothing diagonal lines as shown in Figure C1, so a smooth image can be obtained. .

Note that in the above embodiment, the RO outputs interpolated pixel column data by using the pixel column data of two adjacent lines as address inputs.
Although the configuration is such that an interpolated pixel string is created using M,
Of course, a configuration in which the interpolated pixel array is created by performing arithmetic processing using a CPU or a dedicated processor is also possible.

〔effect〕

As described above, according to the present invention, when reproducing image data in which the scanning density in the sub-scanning direction is set to be higher than that in the main scanning direction, which is often used as a method for reducing the amount of information in facsimile machines, the diagonal line of curve 2 has a stepped shape. to avoid being expressed in
It becomes possible to express it as a smooth curve or diagonal line.

[Brief explanation of the drawing]

Fig. 1 is a schematic diagram showing the principle of the method of the present invention, Fig. 2 is a schematic diagram not showing the same processing method, Fig. 3 is a block diagram showing the circuit configuration for implementing the method of the present invention, and Fig. 4 is a schematic diagram showing the principle of the method of the present invention. FIG. 3 is a schematic diagram of an image reproduced by the circuit. 3...Pixel conversion ROM 11.12...Line memory patent Applicant: Sanyo Electric Co., Ltd. Agent Patent attorney: Noboru Kono My husband's original butterfly. Miσ)ノYri-so

Claims (1)

[Scope of Claims] 1. When reproducing an image signal formed by arranging a plurality of lines of an image signal formed by arranging a plurality of binary pixel signals in one direction in a direction orthogonal to the line direction, two adjacent two Binary pixel signals of a line are compared sequentially between pixels whose line direction positions correspond, and each pixel of one line starts from the pixel next to the first line direction position where both pixels of both lines have one signal value. continuously takes the other signal value, and there are two or three pixels between the pixels of the other line and the second line direction position where the pixels of the other line change to the other signal value. A line of image signals characterized in that a pixel signal of a line interpolated between both lines is created by performing diagonal line smoothing processing by converting the next pixel of one line direction position into the signal value of the other. Interpolation method.