JPH08242364A - Method and device for processing image - Google Patents

Abstract

PURPOSE: To reduce the generation of moire even under a condition that images with different linear density are mixed in a source document by selecting a filter of spatial frequency not in a common multiple relation of detected specific spatial frequency and applying filtering processing to image data obtained after variable magnification processing. CONSTITUTION: Image data obtained after variable magnification processing are inputted to a spatial filter/gradation compensating par 28 to execute filtering processing and a frequency component allowed to generate moire in the image data is extracted from an FFT/filter selecting part 33. A selection part in the FFT/filter selecting part 33 applies a selection signal for specifying a spatial frequency filter having spatial frequency not in a common multiple relation with the detected specific spatial frequency to the compensating part 28. The compensating part 28 selects a spatial frequency filter corresponding to the selection signal to execute filtering processing for the magnification-varied image data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to scaling processing of three primary color signals of red, green and blue obtained by color-separating an original image and reading it.
In particular, the present invention relates to image processing for generating image data that has undergone scaling processing. The image processing of the present invention is performed by, for example, a digital copying machine,
It is applied to printers, facsimile machines, computer image processing, etc.

[0002]

2. Description of the Related Art Generally, in an image forming apparatus such as a digital copying machine, a printer, a facsimile apparatus, a computer image processing apparatus, etc., a laser writing, a CRT display, a liquid crystal display or the like is used for 240 [dpi] to 600 [dpi]
The image is recorded or displayed at the dot density (writing density) of i]. In the image forming apparatus, for example, the original of the image to be recorded is printed lines, specifically 100 lines, 133.
It is a printed matter having a plurality of line densities (lines / inch) such as lines, 150 lines, 175 lines, 200 lines, 250 lines, etc., and the writing density of the image forming apparatus is 400 [dpi], and the magnification ratio of the document to the output image is Consider the case of 50%. In this case, interference fringes (moire) occur in the output image of the 200-line portion of the document. This phenomenon is such that when 200 lines on an original are reduced to 50%, the spatial frequency becomes substantially the same as the writing density of 400 [dpi] of the image forming apparatus, and moire occurs.
In the above description, although the number of lines is 200, both 175 lines and 133 lines generate moire not a little, depending on the magnification.

When the following equation is satisfied, moiré appears remarkably: ρ = κ / M (1) ρ: image recording (or display) density (dpi: number of dots per inch) κ: original line Density (lpi: Number of lines per inch) M: Magnification As an example, the appearance of moire when κ = 175 [lpi] is shown in the graph of FIG. That is, according to the experiments by the present inventors, as shown by the solid line in the graph of FIG.
The point that satisfies the equation (1) is regarded as the worst point of moire generation, and the degree of moire generation gradually decreases as the magnification increases and decreases. Therefore, if a filter effective for smoothing is provided as a spatial frequency filter. The effect of reducing the moire as shown by the broken line in the graph of FIG. 3 was confirmed. Even when the line density of the original image is increased or decreased, the degree of occurrence of moire gradually decreases, and the moire is reduced by providing a filter effective for smoothing as the spatial frequency filter.

Conventionally, in order to deal with such moiré, an image is processed after reading a document or before writing.
The occurrence of moire has been reduced by performing a filtering process on image data using a spatial filter, especially a filter effective for smoothing. A scaling process of image data of a conventional color digital copying machine which takes measures against such moire will be described below with reference to a conventional example shown in FIG. Note that FIG.
The circuit from image reading to recording output ((a) + (b) + (c) in FIG. 5) shown in FIG. 1 is equipped in the digital color copying machine shown in FIG. 1, for example. Therefore, first, the digital color copying machine shown in FIG. 1 will be described.

As shown in FIG. 1, this digital color copying machine is of a one-drum type which is provided with one image forming unit. Generally, a laser printer 1, an image scanner 2 and an automatic document feeder are provided. 3 and 3.

The image scanner 2 has a contact glass 4
The exposure scanning optical system 5 disposed below the table is mechanically moved in the horizontal direction in the figure to perform scanning, and the reflected light image from the original on the contact glass 4 is sequentially formed on the CCD 6 having a one-dimensional line sensor structure. By doing so, the original image is read. More specifically, the reflected light image from the document passes through a lens and the like, and is then separated into R, G, and B according to the wavelength by the dichroic prism 7, and is imaged on the CCD 6 for each color. Thus, the image scanner 2 including the three CCDs 6 can simultaneously detect the image signals of the R, G, and B components for one line in the main scanning direction on the original image. Then, since the exposure scanning optical system 5 subscans,
Two-dimensional image reading is performed on the document.

The laser printer 1 is composed mainly of a single drum-shaped photosensitive member 8. Around the photosensitive member 8, a charging charger 9, a writing unit 10, a developing unit 11 and a transfer drum are formed in accordance with an electrophotographic process. 1
2, the cleaning unit 13 and the like are arranged in order. The writing unit 10 is a well-known unit that uses a semiconductor laser or the like as a writing light source and deflects and scans the surface of the photoconductor 8 by high-speed rotation of a polygon mirror 10a to form an electrostatic latent image. The developing unit 11 selectively performs developing processing for color recording by using toners of different colors of M (magenta), C (cyan), Y (yellow), and BK (black). One developer 1
It is equipped with 1Y, 11M, 11C and 11BK. Also,
The transfer drum 12 holds the supplied transfer paper 14,
The image developed for each color on the photoconductor 8 is subjected to image transfer four times, and a transfer charger 15 is built in at a position facing the photoconductor 8. Such a transfer drum 1
A fixing device 16 that performs a fixing process on the transfer paper 14 that has been peeled off after the transfer process has been completed four times is provided on the second paper discharge side.

Next, the conventional image data processing by the circuit from image reading to recording output shown in FIG. 5 will be described. First, in the image scanner 2 portion shown in FIG. 5A, the signal photoelectrically converted by the three CCDs 6 (analog voltage signal) is amplified by the amplifier 17, and the A / D
A digital value (2
Value) signal. This digital signal is subjected to necessary shading correction (correction of the light amount difference between the left and right of the original light image to make the light amount ratio of the entire surface the same) by a shading correction circuit 19 and then by a line correction circuit 20.
Line-to-line correction of each R, G, B signal is performed. Then, in the scanner γ correction circuit 21, the scanner γ
The gradation of R, G, and B is made uniform, the linear function of the reflectance is converted to the linear function of the brightness, and the interval of each pixel is corrected by the change of the magnification. In the scanner γ dot correction circuit 21, such necessary processing is performed, and R,
As the G and B signals, the IPU (image
Processing / unit) 22.

In the IPU 22 shown in FIG. 5B, R,
The spatial filter 23 for G and B smoothes the halftone image (or sharpens the character image).
I do. Next, in contrast adjustment 24, the intensity of contrast is adjusted, and UCR / UCA processing / color correction /
The color conversion unit 25 performs UCR / UCA processing, color correction, and color conversion processing to convert the R, G, B signals into Y, C, M, Bk signals for writing and recording. The selector 26 selects one color signal from the converted signals in accordance with the image forming timing, and the scaling circuit 27 performs electrical scaling processing in the main scanning direction. The spatial filter / gradation correction unit 28 for writing performs filter processing and gradation processing on the signal after the scaling processing, and outputs the signal to the laser printer 1.

In the laser printer 1 shown in FIG. 5C, the signal (image data) obtained from the IPU 22 is corrected by the printer γ circuit 29 for writing, and then,
It is given to the LD modulation circuit 30 and the LD driver 31, and is used to drive the semiconductor laser (LD) in the writing unit 10. When the LD is modulated and driven based on such image data, optical writing is performed on the photoconductor 8 and an electrostatic latent image is formed.

That is, in such a configuration of the image processing system, the spatial filters 23 for R, G, B and M, C, Y,
The filter processing by the spatial filter / tone correction unit 28 for Bk is said to be useful for reducing the occurrence of moire.

[0012]

The spatial frequency of these spatial filters is set regardless of the magnification and the linear density of the original image. For this reason, moire that may occur due to the relationship between the magnification and the density needs to be adjusted each time so that the spatial frequency becomes appropriate, which is troublesome.

Further, various kinds of linear density images, for example, 133 [lpi], 175 [lpi], and 20 are included in the same document.
It is not possible to deal with the case where images having a linear density such as 0 [lpi] are mixed. There are cases where it is not possible to eliminate moire.

[0014]

According to the present invention, the red, green, and blue primary color signals obtained by color-separating and reading an original image are scaled to a predetermined magnification and separated into a luminance signal and a hue signal. ,
Performs a Fourier transform of the output image data at a specified magnification using the separated luminance signal to detect the specific spatial frequency that causes moire in relation to the writing density, and the common multiple for the detected specific spatial frequency. A filter having a spatial frequency that is out of the function is selected and the image data after the scaling process is filtered.

The image processing apparatus of the present invention which realizes such a method inputs the three primary color signals, and an image reading means for color-separating an original image and outputting it as three primary color signals of red, green and blue. Magnification changing means for changing the magnification to a predetermined magnification and a plurality of types of spatial frequency filters are selectable, and the image data after the magnification changing means is performed by the selected one spatial frequency filter Filtering means for performing filter processing, means for inputting scaled image data to separate into a luminance signal and a hue signal, and output image data at a magnification determined by using the luminance signal separated by this separating means Fourier transform means for detecting the specific spatial frequency that causes moiré in relation to the writing density, and the Fourier frequency of the spatial frequency outside the common multiple relation with respect to the detected specific spatial frequency. Filter selecting means having a selecting means for selecting a filter from the filter processing means.

In the image processing apparatus of one embodiment of the present invention, the plural kinds of spatial frequency filters are filters composed of plural kinds of multi-valued matrix. Further, the image data to be filtered is the luminance signal.

[0017]

According to the image processing method of the present invention, the output image data is Fourier-transformed at a magnification determined by using the luminance signal of the image data subjected to the scaling processing to generate moire in relation to the writing density. Detects a specific spatial frequency, selects a spatial frequency filter that does not become a common multiplication function for this specific spatial frequency, and performs filtering on the image data after scaling, so images with different linear densities on the original Even under a condition in which there are mixed images, the filter processing is appropriate for the image on the document, and the occurrence of moire is reduced.

The image processing apparatus of the present invention brings about the action and effect of the above-mentioned image processing method, and in addition to the conventional image processing function means, has a plurality of types of spatial frequency filters, and a Fourier transform section and a selection section. It suffices to provide filter selection means, and the addition of a simple unit can significantly reduce the occurrence of moire.

In the embodiments described below of the present invention, the spatial frequency filters are 1 × 2, 2 × 1, 2 × 2, 3 × 3, ..., 8 in order to deal with a wide range of specific frequencies that may cause moire. × 8
The spatial frequency filters of different sizes, which consist of multiple binary and multi-valued matrices, are equipped. As a result, the moire reduction effect is high even when the variety of document images is wide. In a later-described embodiment of the present invention, the luminance component of the image data after the scaling processing is also filtered. The amount of data to be processed is R, G, B or Y, M,
Since it is smaller than C, Bk, etc., the filtering process is simple and there is no adverse effect on the hue.

[0020]

【Example】

-First Embodiment- Fig. 2 shows a first embodiment of the present invention. The first embodiment is applied to, for example, a digital full-color copying machine as shown in FIG. 1, and most of its electrical system is configured similarly to the conventional example of FIG. Part of that has been improved.

In the first embodiment shown in FIG. 2, first, the IPU
In 22 the output side of the scaling circuit (magnifying means) 27 is
A circuit 34 for separating a luminance signal and a hue signal is connected as a luminance / hue separating means. The spatial filter / gradation processing unit 28 is connected to the output side of the separation circuit 34 as a filter processing means. The spatial filter / gradation processing unit 28 has a plurality of types of spatial frequency filters that can be selected. These spatial frequency filters have 1 × 2
2x1, 2x2, 3x3, 3x5, 5x5, 6x6, 7
Filters composed of multi-valued matrices having different sizes such as × 7 and 8 × 8 are used. These filter coefficients may be coefficients effective for smoothing. Further, in the gradation correction unit of the spatial filter / gradation correction unit 28, since these spatial frequency filters function as a kind of gradation correction unit, gradation correction processing may be performed separately depending on the selection of these spatial frequency filters. It is supposed to be done.

A circuit 34 for separating a luminance signal and a hue signal is connected to the output side of the scaling circuit 27, and an FFT / filter selection section 33 is provided on the output side of the separation circuit 34.
Are connected as filter selection means. This FFT
The filter selection unit 33 performs a Fourier transform, which is a fast Fourier transform (FFT) on the luminance signal from the luminance / hue signal separation circuit 34, and the spatial filter / level based on the processing result by the Fourier transform unit. The adjustment unit 28 includes a selection unit that selects one spatial frequency filter. The Fourier transform is to detect a specific frequency that can cause moire by extracting a frequency component that can cause moire by performing a fast Fourier transform on target image data. In addition, the selection unit selects one spatial frequency filter having a spatial frequency that is not in a common multiple relationship (out of the common multiple relationship) with respect to the specific frequency detected by the fast Fourier transform so as not to generate moire. It is for.

In such a configuration, the image data read from the document is transferred from the image scanner 2 side to the IPU.
After being input to 22 and appropriately subjected to necessary processing, it is input to the scaling circuit 27 and subjected to scaling processing according to the designated magnification. The image data after the scaling processing is input to the luminance / hue signal separation circuit 34, and only the luminance signal data is input to the FFT / filter selection unit 33. Fast Fourier transform is performed here.

On the other hand, the image data after the scaling process is input to the spatial filter / gradation correction unit 28 and subjected to the filtering process, but moire occurs in the image data from the FFT / filter selection unit. Frequency components are extracted. The frequency component thus extracted is detected as a specific frequency that causes moire in the image data. At this time, since the Fourier transform processing is performed on the luminance component of the image data after the scaling processing, the frequency components that can generate moire due to the relationship between the linear density of the input data, the magnification, and the writing density are It will be detected each time.

The selection unit in the FFT / filter selection unit 33 sends a selection signal designating a spatial frequency filter having a spatial frequency that is not in a public relation to the detected specific spatial frequency, to the spatial filter / level. It is given to the tone correction unit 28. The spatial filter / gradation correction unit 28 selects one corresponding spatial frequency filter according to the selection signal, and performs the filtering process on the corresponding image data after the scaling process. In this filtering process, moire does not occur because the spatial frequency that does not have a common multiple relation to the specific frequency that causes moire for the corresponding image data.

Therefore, according to the first embodiment, even in the case of a copy in which images of different linear densities are mixed on the same document, an appropriate spatial frequency is set for each image area of each linear density. The filter processing is performed by the spatial frequency filter.

-Second Embodiment- A second embodiment of the present invention is shown in FIG. In this second embodiment,
The filtering process is performed only on the luminance signal separated by the luminance / hue signal separation circuit 34. The contrast adjustment 24 of the first embodiment is converted into the luminance / hue signal separation circuit 34, The hue data a * and b * are given to the color conversion / UCR / UCA processing unit 35, and the luminance data L * is UCR after being filtered by the spatial filter 28.
/ UCA processing unit 35. FFT
Similar to the first embodiment, the filter selection unit 33 extracts the frequency component that can cause moire in the image data after the scaling processing by the fast Fourier transform. The selection unit in the FFT / filter selection unit 33 gives the spatial filter 28 a selection signal designating a spatial frequency filter having a spatial frequency that is not in a common multiple relation with the detected specific spatial frequency. The spatial filter 28 selects one corresponding spatial frequency filter according to the selection signal, and performs the filtering process on the luminance data L * after the scaling process. In this filtering process, since the spatial frequency that does not have a common multiple relationship with the specific frequency that causes the moire, the moire hardly occurs.

According to the second embodiment, the recording density signals Y, M, C and K are filtered in the first embodiment, whereas the luminance signal L * is filtered. The amount of filtered data is small and the processing efficiency is high.

The multi-value matrix used as the spatial frequency filter in the spatial filter / gradation correction unit 28 of the first embodiment and the spatial filter 28 of the second embodiment is effective as a multi-value dither matrix for smoothing. With a certain coefficient, it is possible to reduce moire, and at the same time, it is possible to favorably reproduce the gradation of a low density area in an image to be recorded.

[0030]

EFFECTS OF THE INVENTION Even under the condition that images having different linear densities are mixed in the original, the filter processing is appropriate for the image on the original, and the moire is reduced. In addition to the conventional image processing function means, it suffices to provide a plurality of types of spatial frequency filters and a filter selection means having a Fourier transform section and a selection section. The addition of a simple unit significantly reduces the occurrence of moire. can do.

[Brief description of drawings]

FIG. 1 is a block diagram showing an outline of the mechanism of one digital color copying machine in which an image processing apparatus of the present invention can be used.

FIG. 2 is a block diagram showing a configuration of a first exemplary embodiment of the present invention.

FIG. 3 is a graph showing a scaling ratio and a degree of moire generation at a draft line density of 175 lines / inch.

FIG. 4 is a block diagram showing a configuration of a second exemplary embodiment of the present invention.

FIG. 5 is a block diagram showing one of conventional image processing apparatuses.

[Explanation of symbols]

1: Laser printer 2: Image scanner 3: Automatic document feeder 4: Contact glass 5: Exposure scanning optical system 6: CCD 7: Diaclock prism 8: Photoconductor 9: Charging charger 10: Writing unit 10a : Polygon mirror 11: Developing unit 12: Transfer drum 13: Cleaning unit 14: Transfer paper 15: Transfer charger 16: Fixing device 22: IPU (image processing unit)

Claims (4)

[Claims] 1. A red image, a green image obtained by color-separating an original image and reading the original image,
The blue three primary color signals are scaled to a predetermined scale factor, the scaled data is separated into a luminance signal and a hue signal, and the output image data at the scale factor determined using the separated luminance signal Perform a Fourier transform to detect the specific spatial frequency that causes moiré in relation to the writing density, select a spatial frequency filter that is outside the common multiplication function for the detected specific spatial frequency, and select the scaling data. An image processing method characterized by performing filter processing. 2. An image reading means for color-separating an original image and outputting it with three primary color signals of red, green and blue, and a scaling means for inputting the three primary color signals and varying the magnification to a predetermined magnification. A plurality of types of spatial frequency filters are freely selectable, and a filter processing unit that performs a filtering process on the image data that has been subjected to the scaling process by the scaling unit by the selected one spatial frequency filter; A means for inputting image data to separate it into a luminance signal and a hue signal, and a Fourier transform of the output image data at a magnification determined by using the luminance signal separated by this separating means to perform moire in relation to the writing density. A Fourier transform means for detecting a specific spatial frequency that generates a noise and a filter for a spatial frequency that is out of the common multiple relation with respect to the detected specific spatial frequency, are selected from the filter processing means. The image processing apparatus characterized in that it comprises a filter selecting means having a means. 3. The image processing apparatus according to claim 2, wherein the plurality of types of spatial frequency filters are formed of a plurality of types of multi-valued matrices. 4. The image processing apparatus according to claim 2, wherein the image signal to be filtered is a luminance signal.