JPS63222854A - Image processor - Google Patents

Abstract

PURPOSE:To suppress moire pattern, by comparing a dither pattern selected on the basis of a selection signal outputted from a random number processing section with an image signal in a dither processing section and outputting a matrix signal. CONSTITUTION:An image signal 11 is inputted from a signal input section 10 to a dither processing section 20. A pattern output section 30 receives a selection signal 40a for a random number processing section 40 and outputs a selected dither pattern 31 to the dither processing section 20. Levels of an input signal 11 and a signal of the dither pattern 31 corresponding position are compared, and if the level at the side of the input signal 11 is higher, that portion is dither processed as level 1 while it is dither processed as level 0 in other cases so as to obtain a matrix signal 21. When an input signal 12 is fed to the dither processing section 20, the selection signal 40a is switched to feed a dither pattern 32 to the dither processing section 20. When dither processing is made in such a manner, matrix signals 21-23 can be made irregular thereby generation of moire pattern is suppressed, resulting in the improvement of the quality of recorded image.

Description

DETAILED DESCRIPTION OF THE INVENTION "Field of Industrial Application" The present invention relates to an image processing apparatus that processes an image signal that expresses the shading of an image in multiple stages using a so-called dither method.

"Prior Art" Various methods are known for expressing halftone images. A method is also known in which each pixel that makes up an image is expressed individually in multiple stages of gradation, but while this allows for small pixels and high image quality, it is complicated to control and requires special recording paper. There are also problems such as having to use .

On the other hand, a method is also known in which each pixel is composed of a matrix pattern of binary black or white dots, and the proportion of black dots in the pattern is varied to express intermediate tones. . This is characterized in that the apparent density of the pixel as a whole is determined by the number of black dots, that is, the area of the black part.

According to this method, although the pixel area tends to be somewhat large, it is possible to perform digital processing on the individual image signals forming the matrix pattern, which is advantageous in terms of signal processing.

Now, the Didi method is known as a typical method for realizing this method.

In this dithering method, for example, as shown in FIG. 2, image signals obtained by reading an image on a document 1 with an image sensor or the like are manually input to the dither processing section 4 as a set of four 2×2 signals. On the other hand, a dither pattern 3 composed of the same 2×2 pattern as this signal 2 is prepared and manually applied to the dither processing 84. In reality, since the signals input one line at a time are processed in sequence, there is also a method in which the upper half (2 bits) of the human input signal 2 is processed first, and the lower half is processed when the next line is input. It is taken.

Note that this dither processing unit 4 is configured so that each input image signal is “
This is prepared in order to obtain a matrix-like signal 5 that expresses a halftone according to the level of a signal with four density levels from "0" to "3".

The output signal of the dither processing section 4 is the input signal 2 as shown in the figure.
For input signal 2 and dither pattern 3, the signal levels at corresponding positions are compared, and if the input signal 2 is larger, the corresponding position of the matrix signal 5 will be at the "1" level, and in other cases it will be at the "0" level. The image has been binarized.

For example, in the example shown in the figure, the signal level at the upper left corner of the human input signal 2 is "1" and the signal level at the upper left corner of the dither pattern 3 is "0", so the signal at the upper left corner of the matrix signal 5 is "
The signal level at the lower left corner of human input signal 2 is "1", and the signal level at the lower left corner of dither pattern 3 is "1".
” Therefore, the signal at the lower left corner of matrix signal 5 is “0”
becomes.

In this way, when the obtained matrix signal 5 is printed with black dots for level "1" and white dots for level "0", the 2×2 dot matrix 5' allows the level of the input signal to be "1". Can express midtones.

``Problems to be Solved by the Invention'' Now, when performing such dither processing to express a half-tone image, even though the image on the original is of uniform density, the recorded image may Density unevenness with a certain periodicity may occur.

This is a phenomenon called moiré.

To explain using the example in Figure 2, all input signal levels are “1”.
Even if the image is uniform, the recorded image will be a 2 x 2 dot matrix with a repeated pattern with a black dot in the upper left corner.Moiré occurs due to the repetition of this regular pattern. Resulting in.

One possible method for preventing such a phenomenon is to change the dither pattern 3 prepared by the dither processing section 4 as appropriate.

For example, as shown in FIG. 2, the dither pattern 3 used in the previous explanation is rotated by 90 degrees in the direction of arrow 6 to prepare the dither pattern 3' shown in the figure.

This is a pattern completely equivalent to dither pattern 3, and when dither processing is performed using this pattern, even if the position of the black dot changes, the density of the halftone expressed by the dot pattern remains the same.

To create dither pattern 3' from dither pattern 3,
It is sufficient to shift the signal by one bit in a latch circuit or the like that stores the dither pattern. A technique to prevent moiré by performing this 1-bit shift randomly was developed in Japanese Patent Application Laid-Open No. 6.
1-113464.

However, even with this method, some moiré remains. In particular, a 2x2 matrix becomes a 3x3.
As the size increases to 4×4, more effective processing is required.

The present invention has been made with attention to the above points, and an object of the present invention is to provide an image processing apparatus that can sufficiently suppress the occurrence of such moiré.

"Means for Solving the Problems" The image processing device of the present invention includes a signal input section that receives an image signal that represents the shading of an image in multiple stages, and a matrix-like signal that represents the density corresponding to the image signal. a dither processing unit that creates a threshold value for binarizing the image signal; a pattern output unit that generates a plurality of dither patterns arranged in a J square shape; a random number processing section that generates and outputs a selection signal by random number processing, and the dither processing section uses a dither pattern selected by the selection signal output from the random number processing section to It is characterized by comparing the signals and outputting a matrix-like signal.

"Operation" In the above device, a plurality of dither patterns are prepared in the pattern output section, and one of them is selected and outputted by random number processing. If the pattern output section is configured with memory,
All possible equivalent dither patterns are stored in this memory. Then, one of them is randomly selected. In this way, the regularity of the matrix signal obtained by dither processing can be eliminated, and the occurrence of moiré can be suppressed.

Embodiment FIG. 1 is a block diagram showing an embodiment of an image processing apparatus of the present invention.

This device processes image signals 11 to 13 sequentially input from a signal input section 10 to a dither processing section 20 by comparing them with dither patterns 31 to 33 output from a pattern output section 30, and processes the image signals 21 to 23 in matrix form. This is a circuit configured to obtain .

The method of comparing each image signal and the dither pattern in the dither processing section 20 is exactly the same as that described in FIG. 2.

The signal input unit 10 includes an image reading device such as an image line sensor, and a circuit that performs certain processing on the read image signal. Further, the signal input section 10 may be a circuit that supplies an image signal directly created by an information processing device or the like.

In this embodiment, the pattern output section 30 is a read-only
Configure using memory (ROM).

For example, 24 types of dither patterns are stored here, and a random number selection signal 40 generated by the random number processing section 40 is stored.
The configuration is such that any dither pattern can be read by manually inputting a to that address.

As described in the pattern output section 30 of FIG.
With the dither pattern shown in the diagram of the matrix configuration, 24 theoretically equivalent patterns can be created.

That is, by rotating the basic pattern 30a by 90° in the direction of the arrow 6, four types of patterns 30a to 30a'' can be obtained.On the other hand, six types of combinations (30a -f) It is possible. Therefore,
A total of 24 types of dither patterns (6×4), all equivalent, are obtained.

No matter which of these dither patterns is used, a dot pattern with the same proportion of black dots can be obtained for an image signal of the same level. The change in the positions of the black and dots is as explained earlier using FIG. 2.

The apparatus of the present invention having the above configuration operates as follows.

First, a 2×2 array of image signals 11 is input from the signal input unit 10 to the dither processing unit 20 . It is assumed that this signal is obtained by sequentially reading out signals stored in, for example, a two-line buffer memory in this arrangement. On the other hand, the pattern output section 30 receives the selection signal 40a output from the random number processing section 40 and selects one of the prepared dither patterns.
The two dither patterns 31 are output to the dither processing section 20.

The method of comparing the human input signal 11 and the dither pattern 31 is explained in the second section first.
In the same way as explained in the figure, the signal levels at corresponding positions are compared, and if the input signal 11 side is large, that part is set to level "1", and in other cases, dither processing is performed with the level set to "0". In this way, a matrix signal 21 is obtained.

Next, the input signal 12 from the signal input section 10 is sent to the dither processing section 2.
Enter 0. This time, the selection signal 40a input to the pattern output section 30 is switched, and a different dither pattern 32 is input to the dither processing section 20.

In this embodiment, the input signal 11 and the input signal 12 are signals with exactly the same content, but because the dither patterns used are different, the level "1" is placed at a different position from the previously output matrix 21. A matrix-like signal 22 is obtained in which the signals are present.

Of course, the ratio of the number of level "1" and level "0" signals is the same.

Further, when the input signal 13 is input to the dither processing section 20, another dither pattern 33 is outputted from the pattern output section 30 again. As a result, the matrix signal 23 is distorted.

As described above, by randomly selecting one dither pattern from a large number of prepared dither patterns and performing dither processing, the output matrix signals 21 to 23 can be made sufficiently irregular.

"Variation" The image processing apparatus of the present invention is not limited to the above embodiments.

For example, the pattern output section 30 and the random number processing section 40 may be integrated, and this may be constituted by a microprocessor or the like that randomly generates dither patterns using a certain program. Furthermore, moiré can be prevented in exactly the same manner for a dither pattern having a configuration such as 4×4.8×8.

"Effects of the Invention" According to the image processing apparatus of the present invention described above, since dither processing is performed using a dither pattern randomly over a wide range, the occurrence of moiré is sufficiently controlled, and the image quality of recorded images is improved. Can be done.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of an image processing apparatus according to the present invention, and FIG. 2 is a block diagram showing an example of a conventional image processing apparatus. 10... Signal input section, 20... Dither processing section, 30... Pattern output section, 40... Random number processing section. Applicant: Fuji Xerox Co., Ltd. Agent

Claims (1)

[Claims] A signal input unit that receives an image signal that represents the gradation of an image in multiple stages, a dither processing unit that creates a matrix signal to represent the density corresponding to the image signal, and a threshold value for binarizing the image signal. a pattern output section that generates a plurality of dither patterns arranged in a matrix, and a random number processing section that generates and outputs a selection signal for selecting one of the plurality of dither patterns by random number processing, The image processing device is characterized in that the dither processing section uses a dither pattern selected by the selection signal output from the random number processing section, compares the dither pattern with the image signal, and outputs a matrix signal.