JPH0330574A - Pseudo half tone binarization device - Google Patents

Abstract

PURPOSE:To reduce return distortion to be a visual obstacle by continuously connecting a low-pass filter to reduce the energy of a higher frequency area and a band-stop filter, which reduces a frequency spectrum corresponding to the period of a dither matrix, to the output of a analog/digital converting part. CONSTITUTION:The original of dots and lattice patterns is read by a CCD read part 1 of a facsimile equipment. Analog picture data outputted from the CCD read part 1 are inputted to an analog/digital converting part 2 and inputted to a low-pass filter 3 as multiple gradation picture data 8. High-area block picture data 9 outputted from the low-pass filter 3 are inputted to a band-stop filter 4. Band-stop picture data 10 to be outputted from the band-stop filter 4 are successively superimposed with a dither matrix 5 and in a fixed threshold value binarization part 6, pseudo half tone binarizing is executed to these data. Then, the data are outputted to an encoding part 7 and transmitted or recorded as reproduced picture data.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention is applied to a pseudo-halftone binarization device using a systematic dither method for a facsimile machine. In particular, the present invention relates to a moiré removal method when converting a halftone image into pseudo-halftone binarization.

〔overview〕

The present invention uses a pseudo-halftone binarization device that uses a low-pass filter to remove high-order frequencies higher than the reproduction pixel density before superimposing a dither matrix on multi-gradation digital image data and binarizing it. By using a punt stop filter to reduce the frequency spectrum corresponding to the period of the dither matrix size, we were able to reduce aliasing distortion, which was a visual hindrance to humans, and reduce quality deterioration in the low frequency range. It is something.

[Conventional technology]

FIG. 6 is a block diagram of a conventional pseudo halftone binarization device.

Conventionally, a pseudo-halftone binarization device reads a halftone photo original with a CCD reading section 1, and converts the analog image data output from the CCD/CD reading section into multi-tone data using an analog-digital converter 2, as shown in FIG. By periodically superimposing the values of the dither matrix 5 before converting the multi-gradation digital image data into digital image data and binarizing the multi-gradation digital image data in the fixed threshold value binarizing unit 6, a certain specific By changing the ratio of black pixels to white pixels within an area, multi-gradation expression was realized visually by humans.

[Problem that the invention seeks to solve]

However, in such conventional pseudo-halftone binarization devices that use the systematic dithering method of facsimile machines, the original to be read is converted into pseudo-halftone binary images with a certain period, such as a halftone image or a lattice pattern. When converted into values, there is interference between the image with a certain period and the sampling frequency due to the reading pixel density of the CCD reading unit 1, and before the image with a certain period and multi-gradation digital image data are binarized. Two types of interference, one with the period of the dither matrix superimposed on the other, generate aliasing distortion in the frequency domain, that is, moiré, which causes a visually discernible deterioration of image quality in the low frequency domain. was there.

The present invention solves the above-mentioned drawbacks, and aims to provide a pseudo-halftone binarization device that can reduce aliasing distortion, which is a visual hindrance to humans, and reduce quality deterioration in the low frequency region. purpose.

[Means for solving problems]

The present invention includes a reading unit that reads originals with halftone dots and grid patterns, an analog-to-digital conversion unit that digitally converts the halftone image output from the reading unit, and a binarization process by superimposing a dither matrix on digital image data. In the pseudo-halftone binarization device, the pseudo-halftone binarization device includes a fixed threshold binarization means for reducing the energy in a high-order frequency region of the output signal of the analog-to-digital conversion unit, which is considered as the sampling frequency of the reading unit. A low-pass filter and a bandstop filter that reduce a frequency spectrum corresponding to the period of the dither matrix and provide the digital image data to the fixed threshold binarization means are connected in series to the output of the analog-to-digital converter. It is characterized by

[Effect]

The low-pass filter reduces the energy in the higher frequency range of the output signal of the analog-to-digital converter, taken into account as the sampling frequency of the reading section. The bandstop filter reduces the frequency spectrum of the output signal of the analog-to-digital converter corresponding to the period of the dither matrix, and supplies the reduced frequency spectrum as digital image data to the fixed threshold binarization means. The above-described operations can reduce aliasing distortion, which has been a visual hindrance to humans, and reduce quality deterioration in the low frequency region.

〔Example〕

Embodiments of the present invention will be described with reference to the drawings. 1st
The figure is a block diagram of a pseudo-halftone binarization device according to an embodiment of the present invention. In FIG. 1, the pseudo halftone binarization device is a C
CDCD reading unit ratio CCDCD reading unit ratio An analog-to-digital conversion unit 2 that digitally converts a halftone image, and a dither matrix 5 as a fixed threshold binarization means that binarizes digital image data by superimposing a dither matrix.
, adder 21, fixed threshold binarization section 6, and encoding section 7
Equipped with.

Here, the features of the present invention include a low-pass filter 3 that reduces energy in a high-order frequency region considered as the CCDCD reading section ratio sampling frequency of the output signal of the analog-to-digital converter 2; The present invention includes a bandstop filter 4 that reduces the frequency spectrum of the output signal corresponding to the period of the dither matrix and supplies it to the adder 21 as the digital image data.

The operation of the pseudo halftone binarization device having such a configuration will be explained.

FIG. 2 is a characteristic diagram of a low-pass filter in consideration of the sampling frequency of the CCD reading section of the pseudo-halftone binarization device of the present invention. FIG. 3 is a block diagram of the low-pass filter of the pseudo-halftone binarization device of the present invention. FIG. 4 is a characteristic diagram of a bandstop filter in consideration of the period of dithering (IJ) of the pseudo-halftone binarization device of the present invention. FIG. 5 is a block diagram of the bandstop filter of the pseudo halftone binarization device of the present invention.

In the moiré removal method of the present invention, the reading pixel density d in the reading section of the facsimile device satisfies the following relationship with respect to the reproduced pixel density of the pseudo-halftone binarized reproduced image in consideration of the sampling theorem. .

d≧2D (1) A halftone image or lattice pattern containing a certain periodic component is defined as s, (x
, y) and the scanning line signal to be read is 82(x, y), the output image signal p(x, y) is p(x, y)=s, (x, y)・52( x,
y) (2), p (x, y) , s, (
x, y) ・52(X,!>The respective Fourier transforms are P(μ, ν), S, (μ, ν), S2(μ, ν>
Expression (2) is expressed in the form of the convolution product of (3). The reading pixel density d when reading an image containing a certain periodic component is expressed by the formula (
In order to prevent generation of an aliasing distortion spectrum due to the large convolution product of P(μ, ν) in the low frequency region shown in 3), it is desirable to take a value as large as possible. The condition of equation (1) needs to be satisfied for this reason.

Furthermore, if the reproduced pixel density of the pseudo-halftone binarized reproduced image is D, then the frequency spectrum P(μ
, 2) is unnecessary. Since this high-order frequency of 0 or more becomes a cause of distortion in the reproduced image, it is desirable to remove it with a low-pass filter having characteristics around the cut-off frequency.

Next, a dither matrix is superimposed on the multi-gradation digital image data that has undergone the above image signal processing and binarized. However, before superimposing the dither matrix, the dither matrix)
If the spectrum of image data is reduced for the frequency spectrum corresponding to the period of the IJ size, it is possible to prevent the protrusion of a specific frequency spectrum in the low frequency region for P (μ, ν) expressed by equation (3). can.

In Figure 1, first, the CCDCCD of the facsimile machine
The reading unit reads originals with halftone dots and grid patterns. It is assumed that this CCDCD reading unit reads a document at a pixel density of 16 lines/mm. The analog image data outputted from the CCD reading section 1 is inputted to an analog-to-digital conversion section 2, and then inputted to a low-pass filter 3 as multi-gradation digital image data 8. The pixel density of the reproduced image is 8 lines/mm.
Then, the band cutoff characteristic of the low-pass filter 3 is as shown in FIG. In FIG. 3, the count ao of the multiplier 12
Sa+ is realized by a02al=o,5. The transfer function H, (Z) of this filter is H, (Z) = (1 + Z-')/2 (4)
becomes. A simple thinning circuit 13 is provided after the adder 14, and this thinning circuit 13 thins out image data in order to adjust the pixel density of the reproduced image. Since we are now assuming that the read pixel density is 16 lines/mm and the pixel density of the reproduced image is 8 lines/mm, we will assume a circuit that simply thins out each pixel.

High-frequency cutoff image data 9 output from the low-pass filter 3
is manually applied to the bandstop filter 4. however,
Assuming that the matrix size of the dither matrix 5 is 4×4, the pixel density of the high-frequency cutoff image data 9 is 8 lines/mm, so the superposition of the dither matrix has a large spectrum at the frequency of 2 lines/mm. Characteristics can be considered.

Considering the above assumption, it is necessary to construct a band stop filter having the band cutoff characteristics shown in FIG.

FIG. 5 shows the counting of the multiplier 16 using a sixth-order finite impulse response filter (FIR filter) for realizing a bandstop filter. It becomes possible to satisfy the characteristics shown in FIG. 4 by determining each of -b6 to a certain value. The transfer function H2(Z) of this filter is: H2(Z)-Σb,Z-'' (5).

The bandstop image data 10 output from the bandstop filter is sequentially superimposed on the dither matrix 5, converted into pseudo-halftone binarization by the fixed threshold binarization unit 6, and outputted to the encoding unit 7 to be reproduced. transmitted or recorded as data.

〔Effect of the invention〕

As described above, the present invention has an excellent effect of reducing aliasing distortion, which has been a visual hindrance to humans. Therefore, there is an advantage that the disadvantages of the systematic dither method, which is currently the most popular pseudo halftone binarization device, can be reduced.

[Brief explanation of drawings]

FIG. 1 is a block diagram of a pseudo halftone binarization device according to an embodiment of the present invention. FIG. 2 is a characteristic diagram of a low-pass filter in consideration of the sampling frequency of the CCD reading section of the pseudo-halftone binarization device of the present invention. FIG. 3 is a block diagram of the low-pass filter of the pseudo-halftone binarization device of the present invention. FIG. 4 is a characteristic diagram of a bandstop filter in consideration of the period of the dither matrix of the pseudo-halftone binarization device of the present invention. FIG. 5 is a block diagram of the bandstop filter of the pseudo halftone binarization device of the present invention. FIG. 6 is a block diagram of a conventional pseudo-halftone binarization device. DESCRIPTION OF SYMBOLS 1... CCD reading part, 2... Analog-digital conversion tLfE, 3... Low pass filter, 4... Band stop filter, 5... Dither matrix, 6...
・Fixed threshold binarization unit, 7... Encoding unit, 8... Multi-gradation digital image data, 9... High frequency cutoff image data, 1
0... Bandstop image data, 11.15... Delay element, 12.16... Multiplier, 14.17.21.
...Adder, 13...Simple thinning circuit for each pixel of digital image data after passing through a low-pass filter. multi-gradation digital

Claims (1)

[Claims] 1. A reading unit that reads a document with halftone dots and a grid pattern, an analog-to-digital conversion unit that digitally converts the halftone image output from this reading unit, and a dither matrix that superimposes a dither matrix on digital image data. In a pseudo halftone binarization device equipped with a fixed threshold binarization means for binarizing, a high-order frequency region of the output signal of the analog-to-digital conversion unit, which is considered as the sampling frequency of the reading unit, is used. A low-pass filter that reduces energy, and a bandstop filter that reduces a frequency spectrum corresponding to the period of the dither matrix and supplies it to the fixed threshold binarization means as the digital image data are connected in cascade to the output of the analog-to-digital converter. A pseudo-halftone binarization device characterized in that: