JPS6113874A - Picture image processing method - Google Patents

Abstract

PURPOSE:To obtain a reproduced picture image with high quality even if the resolution of an input device is different from the resolution of an output device by sectioning a binary-coded data subjected to dither processing into the unit of prescribed blocks and using a reference table to attain other arrangement. CONSTITUTION:An original picture image subject to raster scanning in an input device is subject to dither processing by a dither matrix and converted into a binary-coded picture and transmitted to a 2-line buffer memory 11 from an input terminal 10, fetched in a register 12 in the unit of 2X2 blocks and becomes input address information to a ROM13 written in the reference table. On the other hand, a horizontal synchronizing signal H-SYNC16 transmitted from the printer side is inputted to an address information counter 15 to designate to which line of the threshold value matrix the scanning line of the printer corresponds. The ROM13 outputs information (4-bit's share) for one line's share of a 4X4 matrix in parallel as an output data and the result is outputted from an output terminal 17. Thus, since the data is converted at each block, the memory capacity is less.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to an image processing method, particularly to an image processing method for obtaining high resolution and high gradation.

[Prior Art] Various methods have been proposed for outputting images with halftones in digital printers and the like.

Examples include a dither method and a density pattern method.

These methods are as follows: (1) An image having halftones can be displayed using a binary display device.

(2) Hardware configuration of the device is easy.

(3) A certain level of image quality can be obtained.

For these reasons, it is widely used in many fields.

Some image output devices using this method require an image memory.

In such a device, the memory capacity of the image memory is as follows:
For example, if an A4 size image is read at 18PEL/am and then converted into binarization using the dither method, it must be possible to produce 210 x 217 x 182 = 18 Mbits per color, and this is especially true for color images. It has to be 4 times larger and has 8MB.

For this reason, there is a method of storing image data in an image memory using a smaller number of PELs than the output PEL number of the printer (how many dots per mm (dat)). That is, for example, suppose an output printer outputs at a dot density of 18 PEL/+am (this is 400 dots per inch, henceforth referred to as 400 dpi (dots per 1 inch)), and input information is output at a dot density of 18 PEL/+am.
m (200 dpi), the memory capacity will be reduced to 1/4 (however, this is only on the premise that the data will be stored after being dithered (binarized)).

1 to 5 show conventional examples of such a method. Assume that a 4×4 dither threshold matrix arranged in pixel units of 8 PEL/mm is shown as shown in Figure 1. Assuming that the input image data is uniform data and given as three data values, an output halftone dot as shown in FIG. 2 will be obtained. In the figure, the cross-hatched areas indicate black output, and the others (white) indicate no output. The output value of the dither is directly used in the second printer with +8PEL/ff1m.
When outputting as shown in the figure, if 2×2 dots of 18PEL/am correspond to one pixel, the correspondence with the input device is 1=
It comes with 1. In the figure, a indicates the size of two dots. However, in this case (Fig. 2), the output halftone dots obtained are 8
×8 There is one for each dot (50 lines/inch), and the roughness is noticeable. Therefore, by applying the halftone dot array in Figure 2 to four 4x4 matrices of actual output dots, as shown in Figure 3, the resulting output will be thinner than the halftone dots in Figure 2 ( 100 lines/inch) Good appearance.

This method is simple in terms of hardware, and involves associating a 4×4 matrix unit (Figure 2) of a two-dimensional array of binary data obtained from an input device with one block, and dividing this into units of four blocks each. (Figure 3).

However, when such a method is applied to the dither method, the following problem occurs.

FIG. 4 shows the result of vertical lines being read by an input device and binarized. If we apply the method described above,
The printer output will be as shown in Figure 5, and will be recorded as if there were two lines.

[Objective] It is an object of the present invention to provide an image processing method that eliminates the above-mentioned drawbacks and can obtain high-quality reproduced images even when the resolution of an input device and the resolution of an output device are different. do.

Another object of the present invention is to realize an image processing method that prevents an increase in memory capacity and makes it possible to store image information in a small memory capacity, and to correct the deterioration in image quality caused thereby.

[Example] Hereinafter, an example of the present invention will be described in detail with reference to the drawings.

FIG. 6 shows a threshold matrix (dither matrix) of an input device according to an embodiment of the present invention. In addition,
It is assumed that this resolution is 8PEL/mm.

Four pieces of data DATAo, 1, 2.3 are now stored in one block. Therefore, the 4×4 shown in Figure 6
Assume that the threshold matrix of is composed of four 2×2 blocks.

Now, if the input data is 3, it is binarized in the input device as shown in FIG. The hatched part in the figure is 1. Others indicate 0. FIG. 8 shows a block diagram of the conversion device in this embodiment.

The original image raster-scanned by the input device is subjected to dither processing using a dither matrix as shown in FIG. 6, and becomes a binarized image as shown in FIG. This binary data is sent from the input terminal (serial data in) 10 to the 2-line buffer memory 11, and is then sent to the above-mentioned DA T
The data is taken into the register 12 in 2×2 block units corresponding to A0~3.

The 4-bit data from the register 12 becomes input address information to the ROM 13 in which the reference table is written. On the other hand, the horizontal direction synchronization signal H-3YNC1e transmitted from the printer side is transmitted from the address information counter 15 (
It is assumed that the printer is LBP and outputs by raster scanning. ) to specify which line of the threshold matrix the printer scan line is. The information counter 15 is constituted by a quaternary counter, and outputs a 2-bit signal as lower address information each time the horizontal synchronization signal H-SYNC1[1 is input. This designated address information is input to the ROM 13 as the lower bits of the address information. The ROM 13 outputs information for one row (4 bits) of a 4×4 matrix in parallel as output data, and this output data is converted into serial data by the parallel/serial converter 14, and the output terminal (
Serial data out) 17.

FIG. 8 shows a converted dot array obtained based on the data in FIG. The obtained dot array was 18PEL/m
m resolution, which is 100 dpi in printing terms.
This shows the halftone dot resolution (when 4×4 is considered as one halftone dot).

No. 1 in FIG. 10 is obtained by converting the data in FIG.
A ROM that provides a conversion table to obtain the output results shown in the figure.
The contents of 13 are shown below. Each division in FIG. 8 is represented by DATA-0, 1, 2, and 3 as shown in the same figure, and each row is L: 0.1,
2.3. In FIG. 7, the values of DATAo, 1 and 2.3 in each proc are 1°0.0.0. Also, the lower address information is 00 as binary data for L=0.1, 2.3,
Give 01°10.11. At this time, 4-bit data of DATA-0, 1, 2, and 3 is obtained as output data of the ROM 13 for each. As a result, the DATA in Figure 7
Information of [1ATA-0, 1, 2, 3 as shown in FIG. 9 corresponding to the information of -0, 1, 2, 3 is obtained.

In this embodiment, data conversion is performed for each block as shown in FIG. 7, so that the memory capacity can be reduced.

FIG. 11 shows binarized input data when there is one vertical line. For example, for a 2×2 matrix in the upper left, FIG. 10 shows the conversion table as NO12. As a result, an output as shown in FIG. 12 is obtained. Therefore, the problem of one vertical line appearing as if it were two vertical lines as shown in FIGS. 4 and 5 is solved. The conversion from Figure 7 to Figure 9 happened to result in the same dot shape as Figure 3 (because the area was assumed constant), but a different pattern was used to achieve linearity as described later. I can do things.

Next, another example will be described.

This example shows the case of multi-valued person and output. It is now assumed that the threshold value matrix of the input device has two threshold values as shown in FIGS. 13(A) and 13(B).

This can be applied when the output device outputs an intermediate density (gray) in addition to white and black, or outputs dots of an intermediate size (area ratio).

Now, suppose that the signals from the input device are obtained as shown in FIGS. 14(A) and (B) (that is, FIG. 14(A),
The data shown in B) is data processed using the dither matrix shown in FIGS. 13(A) and 13(B) when the input image is uniform and the data value is 10, for example.

) The hatched part in the figure is 1. Others indicate 0.

Such an output signal corresponds to the output signal of the register 12 in FIG. 8 having only 3 2 bits for each pixel. The output of register 12 is therefore 8 bits (this is just an increase in the number of bits in the output in the previous embodiment). As a result, converted dot data is obtained, for example, as shown in FIG. 15.

Now, assuming that the output device can output each pixel as three values, R
It is necessary to represent the output data of the OM 13 with 2 bits for each pixel, and at this time, the output of the RO old 3 requires 8 bits of data. Since the resolution of the output pixel is higher than that of the input pixel, the number of output states is greater. Therefore, there is a large degree of freedom in determining the arrangement of output dots from input data. Therefore, for example, two patterns as shown in FIG. 15 and FIG. 16 are obtained. That is, FIG. 14(A),
The pattern shown in FIG. 15 can be generated based on the data in (B), and the pattern shown in FIG. 16 can also be generated. Which pattern to select is based on the characteristics of the output device, and it is sufficient that the input data value and the output density have a linear relationship.

Next, linearization will be explained in detail.

Now, consider ternarizing input data using the threshold matrices shown in FIGS. 13(A) and 13(B).

In FIGS. 13(A) and 13(B), the 2×2 matrix in the upper left is 0, 4. +3.12.18,20,24
．． It can express 28 8 levels of gradation (9 levels including white). The above-mentioned 8 levels of gradation are expressed by a combination of 2-bit values for each pixel in the 2×2 matrix (8 bits in total).

For example, if the number of dot states that can be output is 3 values in a 4 x 4 matrix as shown in Fig.
2=32 (33 including white) level. Therefore, it is possible to freely decide which value of the 32 levels the first 8 levels correspond to.

Figure 17 shows 8
The horizontal axis represents the number of gradations of the level and the vertical axis represents the output density, so that a linear relationship is satisfied.

Also, as shown in Figure 18, for example, by adding gray dots 20 that are not in a straight line relationship in the dot array in Figure 15 (three-value output), it is possible to make them line up on a straight line. .

In the present invention, as described above, different dot arrays are obtained from a group of binarized data obtained by dithering a block by referring to a table in the memory. Although the description is made using a threshold matrix of a specific size for simplicity, this can be applied to any size.

Although the present invention has been described for the purpose of reducing memory costs,
The effectiveness of the present invention is demonstrated even when a signal is read at a resolution of 8 PEL/ram by a fax or the like, and the transmitted signal is outputted by a printer at 16 PEL/mm. This is equivalent to a book that reduces memory capacity in terms of reducing transmission capacity in relation to issues such as communication standards.

[Effects] As explained above, according to the present invention, an image can be processed to obtain an output with high resolution and high gradation. Furthermore, according to the present invention, the memory capacity of the device can be reduced.

[Brief explanation of the drawing]

Figure 1, Figure 6, Figure 13 (A) and Figure 13 (B)
Figures 2 and 3 show examples of dither threshold matrices.
Figure, Figure 4, Figure 5, Figure 7, Figure 8, Figure 11, Figure 1
2, FIG. 14(A), FIG. 14(B), FIG. 15, and FIG. 16 are diagrams showing examples of output halftone dots. Figure 10 is a block diagram showing an example of a device for implementing the present invention, Figure 10 is a diagram showing an example of the contents of a ROM, and Figure 1j is a block diagram showing an example of the contents of a ROM.
FIG. 3 is a diagram showing an example of the relationship between the number of gradations of levels and output density. IO...Input end, 11...Buffer memory, 12...Register, 13...ROM, 14...Parallel/serial converter, 15...Address information counter, 17...Output end . Fig. 1 Fig. 2 Notes Fig. 3 Fig. 4 Fig. 5 Fig. 6 Fig. 71 Fig. 9 Fig. 11 Fig. 12

Claims (1)

[Claims] 1) The output data signal from the input device, which is output at a resolution coarser than the resolution of the output device, is divided into predetermined blocks, and the data arrangement within the block is stored in a table in a storage means. An image processing method characterized by creating another array by referencing. 2) An image processing method according to claim 1, characterized in that data within the block is used as an input address of the storage means. 3) In the image processing method according to claim 1, the storage means is arranged such that the data value of the block obtained from the input device and the density data of the dot recorded by the output device are linear. An image processing method comprising a table configured as follows.