JPS6145675A - Picture element data interpolation system - Google Patents

Abstract

PURPOSE:To obtain a reproducing picture with high quality by making picture element data to be interpolated by using plural small picture elements when picture element data of j line X k row are converted to picture element data of n line X m row. CONSTITUTION:An input 2X2 matrix A is interpolated and converted to a 3X3 matrix A'. Respective picture element data to be interpolated I1-I5 are further resolved to four small picture elements in the main scanning direction, and a level of picture element data to be interpolated is determined by smearing out the four small picture elements suitably based upon original picture element data P1-P4. Thus, first, the picture element data to be interpolated I1 and I2 are determined and next, the picture element data to be interpolated I3-I5 are determined.

Description

[Detailed description of the invention] 〔Technical field〕 The present invention relates to a pixel data interpolation method.

[Prior art]

In recent years, laser beam printers using electrophotography have been used as computer output devices or facsimile terminals. Laser beam printers are expected to have many uses as devices that can quickly convert output signals from computers into high-resolution images. However, in order to output low-resolution pixel data transmitted from a facsimile or the like using a high-resolution laser beam printer, it is necessary to interpolate the data. For example, in order to output facsimile received image data of 8 mm/cm with a laser beam printer of 12 mm/mm, 12 lines are required for both main scanning and sub-scanning.
8 lines = 3/2 times the received pixel data must be interpolated.

Conventionally, such interpolation was performed as follows.

When performing 372 times data interpolation, in the main scanning direction, each pixel ■, ■, ...... as shown in Figure 1(a).
The data is interpolated by dividing into every two pixels and overlapping the second pixel in each division. Similarly, in the sub-scanning direction, as shown in FIG. 1(b), each line Ml, 12 . . . . is divided into every two lines, and data interpolation is performed by overlapping the second line in each division. Such conventional interpolation methods have the disadvantage that image quality is significantly degraded. In other words, the edges of the image become rough because the same pixels or lines overlap regardless of the surrounding pixel data. Furthermore, if conventional interpolation methods are used on halftone images that have been binarized through dither processing, etc., the dot area ratio within the halftone dots will change, the halftone area will change, and the resolution will also drop significantly. There was a drawback.

〔the purpose〕

The present invention has been made in view of the above points, and an object of the present invention is to provide a data interpolation method that makes it possible to obtain high-quality reproduced images.

Another object of the present invention is to interpolate and output pixel data,
An object of the present invention is to provide a pixel data interpolation method that causes less deterioration in image quality.

〔Example〕

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

FIG. 2 is a block diagram of an image processing device to which the present invention can be applied. l is the binarized pixel data 11 II,
2 is a data processing unit that interpolates pixel data, 3 is a transmission unit that transmits the interpolated data, and 4 is an output device such as a laser beam printer that reproduces the interpolated data. The details of the data processing section 2 will be described later.

Next, the principle of the interpolation method according to the present invention will be explained using FIG.

In this embodiment, input pixel data (original pixel data) consisting of a 2 x 2 matrix is interpolated and converted into output pixel data consisting of a 3 x 3 matrix.

For example, if an input image with a resolution of 8 lines/IIm is input with a resolution of 12 lines/IIm,
This conversion is used when outputting with a wood/mm laser beam printer. As shown in FIG. 3(a), the input 2x2 matrix A is interpolated and converted into a 3x3 matrix A. In the figure, P1 to P4 are original pixel data, and matrix
11 to step 5 are interpolated data (interpolated pixel data). In this embodiment, the pixel data to be interpolated is further divided into four small pixels in the main scanning direction, and the level of the element data on the interpolation side is determined by filling these four small pixels as appropriate. In this embodiment, first, interpolated pixel data 11. I2 is determined, and then interpolated pixel data I3. I4. This determines I5. The method for determining interpolated pixel data II will be explained below using FIG. 3(b). Note that the pixel data indicated by diagonal lines is the pixel data on which dots are placed during output, and the width of the diagonal line is the width (size) of the dot. As shown in the figure, when the original pixel data PI and P2 are both "'o" (white), the interpolated pixel data II also becomes "0". Here, the interpolated pixel data "0" means that the four small A state in which all pixels are white is indicated, and interpolated pixel data of "1" indicates a state in which all four small pixels are filled in.

Also, either one of the original pixel data Pi or P2 is '1'
(hatched area), two of the four small pixels of the interpolated pixel data Il are filled out. Also, at this time, the two small pixels closer to the original pixel data at "1'° will be crushed.The original pixel data Pi
, P2 are both "°1", all four small pixels are filled in.

Incidentally, the interpolated pixel data E2 is also determined by the original pixel data P3 and P4 in the same manner as described above.

Next, a method for determining interpolated pixel data 3 will be explained using FIG. 3(c). As shown in the figure, when both the original pixel data Pi and P3 are "0", the interpolated pixel data also becomes "0". Also original pixel data PI.

When either P3 is "1", the interpolated pixel data alternately fills in two of the four small pixels. This is to prevent the black of the interpolated pixel data from shifting. Also, original pixel data PI
, P3 are both °l'', all four small pixels are filled in.

Incidentally, the interpolated pixel data E4 is also the original pixel data P2,
Determined by P4.

Next, the method for determining the interpolated pixel data I5 is shown in FIG.
This will be explained using e). The interpolated pixel data E5 is determined by the interpolated pixel data II and I2 as shown in the figure. Also, If and I2 are both “O”, and either Il or I2 is °゛l”°, If.

The interpolated pixel data I5 in the case where both the angle 2 and the angle 2 are 1° is determined in the same manner as described in FIG. 83(C).

FIG. 5 shows an example in which input pixel data consisting of a 2×2 matrix is converted into output pixel data consisting of a 3×3 matrix using the interpolation method described above. Note that the interpolated 2×
There are 16 types of 2 matrices (input pixel data) as shown in FIG. 4, and each of the matrices 1 to 0 in FIG. 4 corresponds to each of the matrices 2 to 0 in FIG. 5.

In addition, instead of the 3X3 matrix shown in Fig. 5, the 3X3 matrix shown in Fig. 6 (
The 3×3 matrix shown in b) may be used as output pixel data. The matrices 1 to 0 in FIG. 6 also correspond to the matrices 2 to 0 in FIG.

The output pixel data shown in FIG. 6(b) is
This is effective in the case of an apparatus that records dots by pulse width modulation as shown in Figure (a).

At this time, in order to faithfully reproduce the output pixel data shown in FIG. 6(b), it is necessary to change the timing of pulse width modulation as shown in FIG. 6(a).

The area ratio between white and black of the output pixel data shown in FIGS. 5 and 6(b) is the same when compared with the input pixel data of FIG. 4, so the interpolation method as described above can be used. This can prevent deterioration in image quality during interpolation.

Furthermore, even when the input pixel data is a binary image that has been dithered using a threshold value matrix that is an integer multiple of 2×2, fluctuations in the halftone level can be prevented by performing the above-described interpolation.

In this embodiment, interpolated pixel data is output by a laser beam printer, so by making the laser beam elliptical and performing pulse width modulation as shown in Figure 7, an excellent reproduced image can be obtained. In this embodiment, the ratio of the beam diameter in the main scanning direction to the sub-scanning direction is set to 1:4 as shown in the figure.

Next, the details of the data processing section 2 which interpolates the input pixel data as described above are shown in FIG. 8. The numerical value of each line indicates the number of bits of data. What is 11 in the diagram? Input terminal for inputting valued pixel data, 1
Reference numeral 2.13 is a line memory that alternately stores 1247 main scanning pixel data, and 14 is a main scanning interpolation control unit, which controls the operation timing of the line memory, a converter, etc. to be described later. The pixel data stored in the line memory 12 or the line memory 3 is sequentially read out by the main scanning interpolation control section 14, and converted into a signal every 2 bits by the 2-bit parallel converter 15. This signal is a parallel binary signal of two pixels. In the first pixel data converter 16, the pixel data is converted (interpolated) as shown in FIG. 3(b). In other words, the first pixel data conversion unit 16 inputs parallel binary signals of two pixels and outputs a digital signal of two bits per pixel in parallel for three pixels (there are four types of interpolated pixel data, so two bits are required for each pixel). ).

The parallel-to-serial converter 17 serially outputs the three input pixels to complete the interpolation in the main scanning direction.

The first pixel data conversion unit 16 is constituted by a ROM or the like, and a conversion table for performing conversion as shown in FIG. 3(b) is stored in advance in this memory. The 2-bit pixel data that has been interpolated in the main scanning direction is:

Line memory 18.19 or 20.2 for 2 lines each
It is stored in 1. Reference numeral 12 denotes a sub-scanning interpolation control section, which controls the operation timing of a line memory, a conversion section, etc. to be described later. Line memory 18°19 or line memory 20
．． The 2-bit pixel data stored in 21 is sequentially read out in parallel for 2 pixels by the sub-scanning interpolation control section, and is input to the second pixel data conversion section 23. The second pixel data conversion unit 23 performs interpolation in the sub-scanning direction using two pixels (each pixel is 2-bit data) in the sub-scanning direction. As shown in FIGS. 3(c), (d), and (e), there are, for example, 12 ways of interpolation in the sub-scanning direction, so the pixel data converter 23 converts a 2-bit pixel signal into a 4-bit pixel signal. . At this time, the second pixel data converter 23 simultaneously outputs 4-bit pixel signals for three pixels in the sub-scanning direction. The 1-pixel data conversion section 23 is constituted by a ROM or the like, and a conversion table for performing conversions as shown in FIGS. 3(C), (d), and (e) is stored in advance in this memory. Therefore, in the circuit shown in FIG. 8, three lines of data are converted while two lines of data are input to the line memory 18.19 or 20.21. The data selector section 24 selects the second pixel data.
This is for selecting each line from the converter 23. 25 is a pattern generation section, and a data selector section 24
Pulse width modulation is performed in accordance with the 4-bit output signal from the 4-bit output signal, and an interpolated pixel signal is output to the output terminal 16. This output terminal is connected to a laser beam printer 4 shown in FIG. 2, and the laser beam printer 4 reproduces a high-quality image on recording paper in accordance with the pixel signals.

〔effect〕

As described in detail above, according to the present invention, deterioration in image quality when pixel data is interpolated can be prevented.

Further, according to the present invention, even when interpolating dithered binary data, halftones can be faithfully reproduced without changing the dot area ratio within the halftone dots.

[Brief explanation of the drawing]

Figures 1 (a) and (b) are diagrams for explaining the conventional interpolation method, Figure 2 is a block diagram of an image processing device to which the present invention can be applied, and Figures 3 (a) to (e) are diagrams for explaining the conventional interpolation method. A diagram for explaining the interpolation method according to the invention, FIG. 4 is a diagram showing each pattern of input pixel data, FIG. 5 is a diagram showing each pattern of interpolated output pixel data, and FIG. 6 (a) is a diagram showing pulse width. A diagram showing an example of output by modulation, FIG. 6(b) is a diagram showing another pattern example of interpolated output pixel data, FIG. 7 is a diagram showing a state in which the laser beam is shaped into an ellipse, and FIG. 8 is a diagram showing data processing. FIG. Here, 2 is a data processing unit, 4 is a laser beam printer,
12.13 is a line memory, 15 is a 2-bit parallel converter, 16 is a pixel data first converter, 17 is a parallel-to-serial converter, 18 to 21 are line memories, 23 is a pixel data second converter, and 24 is a pixel data second converter. The data selector section 25 is a pattern generation section. Patent applicant: Nippon Telegraph and Telephone Public Corporation Canon Co., Ltd. (α) AA・ (d) =, > Q <> 9 tc) 9

Claims (1)

[Claims] Divide the input pixel data into a matrix consisting of j rows x k columns (j and k are both natural numbers), and interpolate the j rows x k column pixel data to create n rows x m columns (n and m are both natural numbers). Natural number)
1. A pixel data interpolation method for forming a matrix of pixel data, wherein the pixel data to be interpolated among the pixel data of the matrix of n rows by m columns consists of a plurality of small pixels.