JPS61150574A - Halftone recording device - Google Patents

Abstract

PURPOSE:To enable to make a high grade halftone picture at relatively low cost by providing plural gradation expressing media etc. such ass preparing the plural sizes of dot matrix etc., and selecting and using these properly according to the density distribution of a picture. CONSTITUTION:Input picture signals 21 are signals obtained by the plane scanning of an original or a subject. The signals are inputted to a character halftone discriminating circuit 22 and a density distribution discriminating circuit 23 and the discrimination is made as to whether it is a character area or halftone area and whether it is an area for which the resolution is required or an area for which the smoothness is required. Input picture signals 21 are delayed by specified amount by a delay circuit and supplied to three processing circuits 27-29, and subjected to different picture signal processing respectively. Signals for recording 35-37 processed by processing circuits 27-29 are inputted in parallel to a selector 26, and the most suitable one is selected by the result of a discriminating information 24, 25, and sent out to a recording section 39 as a recording signal 38.

Description

DETAILED DESCRIPTION OF THE INVENTION "Field of Industrial Application" The present invention relates to a halftone recording device that can digitally perform image processing to express halftones.

``Prior Art'' Thermal transfer recording type printers, which use a thermal head to transfer thermal transfer ink applied to a thermal recording medium onto recording paper, are used as recording devices that perform digital image processing, such as facsimile machines and word processors. Widely used. Inkjet recording methods and laser xerography, which uses laser beams for recording, are also used in various fields as recording devices that perform digital recording.

In such devices that perform digital recording, the number of gradations that can be expressed by one pixel is only two levels (so-called white and black) or only a few levels. Therefore, it has been substantially difficult to express gradations for each pixel.

Therefore, various techniques have been proposed in which a predetermined number of pixel groups are set as the minimum unit of gradation expression, and this is used to express many gradations. Normally, the minimum unit for gradation expression (hereinafter referred to as gradation expression unit) is constituted by a dot matrix (IJ sock) consisting of a plurality of pixels. The dither method and the density pattern method are commonly used as methods for expressing gradations using these methods.

By the way, when evaluating a halftone image using a gradation expression unit, two different criteria, such as resolution and graininess, are generally used. Here, resolution refers to the total number of distinguishable white and black stripes in the time direction of an image. In order to increase the resolution, it is necessary to define a dot matrix as small as possible as a gradation expression unit to express the image. Therefore, in order to increase the resolution, it is necessary to arrange the dots so that many gradations can be expressed with a small number of dots. Such a technique is disclosed, for example, in Japanese Patent Application No. 106649/1982. In this patent application, by changing the arrangement of printed dots in the dot matrix, the area of the printed portion in the matrix is changed, thereby expressing many gradations.

Next, graininess, which is another criterion used in the evaluation of halftone images, refers to the appearance of an image in a rough state due to the granular structure constituting the image. For example, a person's skin needs to be expressed smoothly, so it is desirable that the skin has little graininess. As a standard for evaluating graininess, Roger Dawley (Roge
There is the following equation proposed by R.Dooley.

= 1.80"ml 7 sheets, VTF (jyu,) G = 1.2
7・Δu'e ......(1) Here, WS is the Wiener spectrum6, and in normal measurement, it is the concentration fluctuation data ΔD' (X) obtained by scanning with a microdensitosorter. It can be calculated using the following formula.

however However, p-q is the width and length of the slit.

Further, D* is the average density, and VTF is the spatial frequency characteristic of the visual system, which is approximated by the following equation.

However, when Δu<l cycles/mm, VTF (
j Δu)=1 Further, ΔU is the step interval of the spatial frequency.

In order to make the graininess less noticeable, it is necessary to arrange the dots regularly. Therefore, in order to express many gradations using this method, it is necessary to make the dot matrix as large as possible.

``Problem to be Solved by the Invention'' In other words, in order to increase resolution, it is better to make the dot matrix or gradation expression unit as small as possible, and this causes deterioration of graininess. On the other hand, if you try to improve the graininess, the dot matrix or gradation expression unit becomes larger and the resolution deteriorates.

Of course, if the density of the dots constituting the recorded image is set sufficiently high, both resolution and graininess can be satisfied. However, this naturally has its limits in terms of device cost and recording speed, and no significant improvement could be expected.

In view of these circumstances, an object of the present invention is to provide a halftone recording device that can satisfy both resolution and graininess.

"Means for Solving the Problem" In the present invention, as shown in principle in FIG. A gradation expression medium etc. specifying means 12 that specifies the type of expression medium and parameters, an image processing means 13 that performs image processing using the specified gradation expression medium and parameters, and The halftone recording apparatus is provided with a recording section 14 for recording a recorded image. An example of a gradation expression medium is to prepare dot matrices of different sizes.

According to the present invention, for image parts such as skin parts that do not require resolution, the size of the halftone expression unit is increased, for example, to satisfy the graininess requirement. Furthermore, for portions such as line drawings that do not require smoothness, the size of the halftone expression unit is reduced to satisfy the resolution requirement.

Since image processing is performed using the gradation expression medium selected in this way, conflicting demands regarding image quality can be satisfied.

“Example j The present invention will be described in detail with reference to Examples below.

FIG. 2 shows the main parts of the halftone recording apparatus of this embodiment. An input image signal 21 read by a reading device (not shown) and A/D converted is supplied to this device. The input image signal 21 is a signal obtained by scanning a document or a subject in a plane, and is input to a character/halftone discrimination circuit 22 and a density distribution discrimination circuit 23 to (i) discriminate whether it is a character area or a halftone area, and (ii) It is determined whether an area requiring resolution is an area requiring smoothness. Discrimination result information 24 and 25 of these discriminator circuits 22 and 23 is supplied to the selector 26.

After the input image signal 21 is delayed by a predetermined amount in a delay circuit (not shown), it is also supplied to three processing circuits 27 to 29.
Each undergoes different image signal processing. Of these, the simple multi-value processing circuit 27 performs gradation expression on a pixel-by-pixel basis based on the input image signal 21, and can reproduce characters or line drawings well. In this embodiment, the ternary dither processing circuit 28 uses a 3.times.3 dot matrix as shown in FIG. 3, and expresses the density of each output dot 31 in ternary values. The dot matrix 32 processed by the ternary dither processing circuit 28 is a submatrix (described next).
Its size is smaller than that of the dot matrix created by the IJ Fuchs logic circuit 29, and a halftone image can be reproduced with high resolution.

The submatrix IJ Fuchs logic circuit 29 used in this embodiment is a circuit for expressing one gradation using a 6×6 dot matrix, as shown in FIG. This dot matrix 33 is composed of four 3×3 sub-matrices 34A to 34D.

The dot matrix 33 processed by the submatrix processing circuit 29 makes the quantization levels in the two diagonally located submatrices 34A and 34C larger (or smaller) than those in the other submatrices 34B and 34D. This allows for smooth reproduction of halftones. Regarding this, for example, B. E. Bayer (B. E. Bayer)
OPTIMUM M8THOD FORTWO-LBV
BL REN-DITION OF C0NTINUO
US-TONBPICTURBS” (1888197
Imitation halftone pattern (Imitation halftone patt)
ern).

Recording signal 35 after being processed by each processing circuit 27 to 29
-37 are input in parallel to the selector 26, and the optimum one is selected based on the above-mentioned discrimination result information 24.25,
It will be sent to the recording section 39 as a recording signal 38.

In such a halftone recording device, the character/halftone discrimination circuit 22 that distinguishes between characters and halftones will be explained first. Fifth
The figure shows an example of this discrimination circuit. An input image signal 21 corresponding to one pixel is sequentially input to an input terminal 41° of this circuit in synchronization with a video clock (not shown). The input image signal 21 is supplied to a first line memory 42, and is also supplied to first to third latches 43. which sequentially latch the input image signal 21 consisting of parallel predetermined bits of digital data one pixel at a time. ~433 and latched.

On the other hand, the input image signal 21 supplied to the first line memory 42 is delayed by a time corresponding to one line of the image read on the reading side by line sequential scanning, and is then supplied to the second line memory 44. In addition, the first to third latches 4
3. - 433 are sequentially delayed by one pixel and supplied to the fourth, zeroth, and fifth latches 434, 43o, and 435, which have the same configuration. The image signal supplied to the second line memory 44 is also read out after being delayed by a time corresponding to one line, and the sixth to eighth latches 436 to 436 having the same configuration as described above are read out.
43. are sequentially delayed by one pixel and supplied.

The character/halftone discrimination section 46 is constituted by a ROM ('J-only memory), and inputs the latch outputs of the 0th to 8th latches 43o to 438 in parallel.
The determination result information 24 is read out using this as address information. The nine types of latch outputs used as address information here are a 3×3 pixel group P whose image state is to be determined, as shown in FIG. - their concentration levels d for P8. This is an information group representing -d8. Nafu pixel P. is a pixel that plays a central role in determining the image state, and will hereinafter be referred to as a pixel of interest in this specification.

Now, the density level d of each pixel. In order to distinguish between characters or halftones using the information group representing -d8, this character/
The halftone discrimination section 46 examines the dispersion of density levels. variance σ
can be expressed by the following formula.

σ−7r, (d, do)2 By comparing the variance σ obtained in this way with a predetermined threshold lTH, the 3×3
pixel group P. - It is possible to determine whether P8 belongs to a character image or a halftone image. This relationship is as follows.

(i) σ>ftu...Character image (ii) σ≦
1, □... The principle of discrimination of halftone images and above will be explained with reference to FIGS. 7 and 8. First, FIG. 7 shows the density distribution of pixels in a character image portion of a document, that is, an image portion composed of characters and line drawings.

In a character image, there is a peak corresponding to the ground color (background) of the document in a relatively low-density part, and a peak corresponding to a so-called black part of ink or the like is in a relatively high-density part. On the other hand, FIG. 8 shows a halftone image portion, that is, an image portion expressed by photography or halftone dot printing, and a uniform density distribution is obtained. Due to this difference in concentration distribution,
If the variance σ is large, the area has a strong tendency to be a character image, and conversely, if the variance σ is small, the area has a strong tendency to be a halftone image. The threshold value Rr□ is set to an optimal value by comprehensively taking into consideration the background color of the document and the density of the ink used. The character/halftone discrimination section 46 stores the density level d of each pixel. -d, is stored in advance, and 1-bit discrimination data representing whether or not it is a character image is sequentially output corresponding to the pixel of interest.

On the other hand, the density distribution determination circuit 23 examines changes in optical density around the pixel of interest and determines whether high resolution is required.

In general, there are two types of images: one in which the optical density changes rapidly as shown in FIG. 9, and the other in which the change appears gradually as shown in FIG. 10. In the former case, graininess is hardly a problem, and rather high resolution is required to obtain high image quality. In the latter case, smoothness is important, for example when expressing human skin, and there is little requirement for resolution.

As shown in FIG. 11, the density distribution discriminating circuit 23 includes a smoothing processing section 51 and a density change discriminating section 52. The smoothing processing unit 51 is composed of a low-pass filter for smoothing a portion where the density level changes rapidly, and is used by the character/halftone discrimination circuit 22 to discriminate between a character portion and a halftone portion. This is for suppressing the content of the input image signal 21 below the threshold value ITH.

The density change discrimination section 52 basically has the same configuration as the character/halftone discrimination circuit 22, except that the character/halftone discrimination section 46 in FIG. 5 is replaced with a density change discrimination section.

The density change determining unit 52 examines the density dispersion within the region of interest after the high frequency component has been cut, and determines whether this exceeds a set threshold value j2TH'. The difference between the maximum and minimum concentrations may be compared with a predetermined threshold. When the density change has occurred rapidly to some extent, the density change determination unit 52 outputs a signal requesting high resolution as the determination result information 25. In other cases, a signal indicating that high resolution is not required is output as the determination result information 25.

The selector 26 selectively selects three types of recording signals 35 to 37 from the two types of discrimination result information 26 as follows, and outputs the selected signals as a recording signal 38.

(i) When the determination result information 24 is determined to be a character image.

Recording signal 35° (ii) by simple multivalue processing circuit 27
When the determination result information 24 determines that the image is a halftone image, and the other determination result information 25 determines that a high resolution is required.

The recording signal 36° (iii
) When the determination result information 24 determines that the image is a halftone image, and the other determination result information 25 determines that high resolution is not required.

Recording signals 37 by the submatrix processing circuit 29 As already explained, each of the recording signals 35 to 37 is selected because it is suitable for expressing characters, high-resolution halftones, or halftones with little graininess. The recorded recording signal 38 allows the recording unit 39 to perform good recording. The switching operation of the selector 26 may be performed for each pixel of interest, that is, for each pixel, or may be performed for each 6×6 dot matrix.

In the halftone recording apparatus of the embodiment described above, after distinguishing between character images and other halftone images, the halftone images are divided into two in accordance with both resolution and smoothness requirements, and different image processing is performed. However, the present invention is not limited to this, and it is of course possible to separate and process the input image so as to meet both requirements for resolution and smoothness. Further, although monochrome recording has been described in the embodiment, the present invention can also be applied to signal processing for each color in two-color or multi-color recording. Further, in the embodiment, the density distribution was determined based on the degree of density change, but it is also possible to use other methods such as analyzing the spatial frequency of the image.

"Effects of the Invention" As described above, according to the present invention, a plurality of gradation expression media, etc., such as a plurality of IJ sock sizes are prepared, and these are appropriately selected and used according to the density distribution of the image. Therefore, high-quality halftone images can be created at relatively low cost.

[Brief explanation of drawings]

Figure 1 is a block diagram showing the principle of the present invention, Figures 2 to 1
Figure 1 shows an embodiment of the present invention, of which Figure 2 is a schematic configuration diagram of a halftone recording device, Figure 3 is a configuration diagram of a matrix used in a ternary dither processing circuit, and Figure 4 is a sub-matrix diagram. Block diagram of the matrix used in the processing circuit, No. 5
The figure is a block diagram of a character/halftone discrimination circuit, Figure 6 is an explanatory diagram showing the planar arrangement of pixels used as discrimination data in this discrimination circuit, Figure 7 is a density characteristic diagram of a character image,
Figure 8 is a density characteristic diagram of a halftone image, Figure 9 is a density characteristic diagram of an image portion that requires high resolution, Figure 10 is a density characteristic diagram of an image that requires smoothness, and Figure 11 is a density characteristic diagram of an image that requires smoothness. FIG. 2 is a block diagram of a distribution discrimination circuit. 11... Density distribution determining means, 12... Gradation expression medium specifying means, 13...
...Image processing means, 14.39...Recording unit, 23...Density distribution discrimination circuit, 26...Selector, 27...Simple multivalue processing Circuit, 28... Three-value dither processing circuit, 29...
・Subma) IJ Fuchs logic circuit. Applicant: Fuji Xerox Co., Ltd. Representative
Patent Attorney Ume Yu Yamauchi 1st
Fig. 2 A Fig. 6 Fig. 8 Fig. + Ft store record Fig. 10 Scanning 75 and Fig. 7 Fig. 9 Quantitative 75 songs 232'

Claims (1)

[Claims] 1. In an apparatus for creating a recorded image by decomposing an input image into pixels, there is provided a density distribution determining means for determining the density distribution of the input image, and a gradation expression method according to the determined density distribution. A gradation expression medium etc. specifying means for specifying the type of medium and parameters, an image processing means for performing image processing using the specified gradation expression medium etc., and a recording section for recording the image processed by the image processing means. A halftone recording device comprising: 2. The gradation expression medium etc. specifying means selects and specifies one of the dot matrices of different sizes prepared for gradation expression according to the density distribution of the input image. A halftone recording device according to claim 1.