JPH10276328A - Image processing unit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To interlock various processing functions with revision of a density notch resulting in matching a density fluctuation range of a reproduced image with a sense of the user. SOLUTION: The processing unit is provided with a CCD 1, an A/D converter 2, a shading correction section 3, an MTF correction section 4, a main scanning magnification section 5, a gamma conversion circuit 6, a binarization section 7, a laser diode(LD) control section 8, an LD 9, an operation section 10, and a control section 11. Then the control section 11 controls each unit to execute operation contents to revise simultaneously the MTF correction characteristic, the gamma characteristic and the binarization characteristic in interlocking with a density notch instructed by the operation section 10.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital image processing apparatus for reading an image, for example, a digital copying machine,
The present invention relates to a facsimile apparatus, and relates to an image processing apparatus capable of easily providing desired image quality by linking various functions to a density notch.

[0002]

2. Description of the Related Art In a conventional digital image processing apparatus, parameters are tuned so that an optimum image can be reproduced for each mode. The reproduction of the halftone can be set as desired by the user, and the selection means is also easy,
Alternatively, the image quality is improved by determining the characteristics of the image area and selecting a desired filter processing. For example, JP-A-5-145759, JP-A-6-62252, and JP-A-6-70168 disclose the contents related to the above.

[0003]

As described above, the smoothing filter processing result is output to the halftone dot area, and the original image data and the edge enhancement filter processing result are mixed and processed in the other areas. It has been conventionally proposed to achieve desirable filter processing and image quality improvement, or to enable a user to easily select a desired intermediate gradation characteristic.

[0004] However, even if the image of the standard density is optimized in these apparatuses, it is often difficult to easily reproduce the “dark” or “light” that the user desires. This is because the level at which the user feels dark or light differs depending on the user.

The present invention has been made in view of such a background, and an object of the present invention is to link various processing functions to a change in a density notch so that a density variation range of a reproduced image matches a user's sense. It is an object of the present invention to provide an image processing apparatus capable of causing the image processing apparatus to execute the processing.

[0006]

In order to achieve the above object, a first means is an image reading means for optically reading an image and converting it to an electric signal, and an A / D for converting an analog signal to a digital signal. Conversion means, shading correction means for correcting density distortion of the read image, MTF correction means for correcting MTF degraded during optical reading,
Main scanning magnification means for electrically enlarging / reducing in the main scanning direction, and γ for performing density conversion in consideration of density characteristics of a reading system, process characteristics of a writing system, and variable characteristics of a density notch.
Converting means, binarizing means for converting a multi-level image signal into a binary signal, LD controlling means for controlling driving of a writing laser diode, optical writing means for forming a latent image on a photoconductor, Image forming means for forming an image on a sheet of paper, operating means for inputting an operation mode of processing, and controlling each unit to execute the operation content, and interlocking with a density notch on the operating means. Control means for simultaneously changing the MTF correction characteristic, the γ characteristic and the binarization characteristic.

The second means includes an image reading means for optically reading an image and converting it into an electric signal, an A / D converting means for converting an analog signal into a digital signal, and a shading for correcting density distortion of the read image. Correction means, smoothing means for removing moire in the read image, main scanning magnification means for electrically enlarging / reducing in the main scanning direction, density characteristics of the reading system, process characteristics of the writing system, density notch Γ conversion means for performing density conversion in consideration of the variable characteristics of the above, dither processing means for simulating the gradation of a multi-valued image signal, LD control means for controlling the drive of a writing laser diode, Optical writing means for creating a latent image on the body, image creating means for creating a latent image on paper, operation means for inputting an operation mode of processing,
Control means for controlling each unit to execute the operation content and simultaneously changing a smoothing filter characteristic, a γ characteristic and a dither characteristic in conjunction with a density notch on the operation means.

The third means includes an image reading means for optically reading an image and converting it to an electric signal, an A / D converting means for converting an analog signal to a digital signal, and a shading for correcting a density distortion of the read image. Correction means, smoothing means for removing moire in a read image, and MTF for correcting MTF degraded during optical reading
Correction means, main scanning magnification means for electrically enlarging / reducing in the main scanning direction, and gamma conversion for performing density conversion in consideration of density characteristics of a reading system, process characteristics of a writing system, and variable characteristics of a density notch. Means, gradation processing means for performing error diffusion processing of the multi-valued image signal, phase control means for generating a phase signal for dot formation at the time of writing, LD control means for controlling the driving of the writing laser diode, Optical writing means for creating a latent image on a photoreceptor, image creating means for creating a latent image on paper, operation means for inputting an operation mode of processing, and each unit for executing operation contents And a control means for simultaneously changing the smoothing filter characteristic, the MTF correction characteristic, the γ characteristic, the gradation characteristic, and the phase characteristic in conjunction with the density notch on the operation means.

The first means is to reproduce a pencil original at a high density only by setting the density notch to the high density side in the operation mode called the binary character mode, and to skip the background density by setting the density notch to the low density side. Is to reproduce.

The second means is to mix a density notch group for optimizing reproduction of a silver halide photograph and a density notch group for optimizing reproduction of a halftone print original in an operation mode called a binary photo mode. An original having different characteristics is optimally reproduced only by a simple density notch operation.

The third means is a notch group for sharply and densely reproducing a line image mainly in a character portion and a notch for softly and thinly reproducing a halftone mainly in a picture portion in an operation mode called a multi-value character / photo mode. By mixing the groups, a reproduction density suitable for the user's feeling is obtained.

[0012]

Embodiments of the present invention will be described below with reference to the accompanying drawings. FIG. 1 is an overall block diagram of an image processing apparatus according to a first embodiment of the present invention. This image processing device has a configuration for realizing a density variation in a binary character mode, a variation coefficient that matches a user's feeling by linking a filter coefficient, a γ table, and a binarization threshold for each density notch. Has become. In other words, when a dark density notch is selected, a faint portion of a pencil character is clearly reproduced, and when a light density notch is selected, a mechanism is provided to delete unnecessary images due to background stains of a document or dust attached to a reading system.

The image processing apparatus shown in FIG.
/ D converter 2, shading correction unit 3, MTF correction unit 4, main scanning scaling unit 5, γ conversion unit 6, binarization unit 7, LD
(Laser diode) control unit 8, LD 9, operation unit 10,
The control unit 11 is provided.

Hereinafter, the operation will be described. A document on a document table (not shown) is irradiated with light from a lamp to read an image line by line. The read optical signal is input to the CCD 1 for each line, and converts the optical signal into an electric signal. The output signal from the CCD 1 is converted in the A / D converter 2 into, for example, an 8-bit digital signal. Although the number of quantization steps is not limited to 8 bits, in the present embodiment, quantization is performed to 8 bits.

To correct uneven illuminance due to the position of the optical reading system, shading correction is performed by the shading correction unit 3. The shading correction reads an absolute white reference white board and stores the data in a line memory. The original read image data of each line is normalized by the reference data stored in the line memory.

With respect to the data after shading correction, since the MTF deterioration in the lens / optical system is corrected by the MTF correction unit 4, the sharpness of the edge portion is guaranteed by the enhancement filter. With respect to the MTF-corrected image data, the main scan conversion unit 5 performs enlargement / reduction electric scaling.

In the case of the same magnification, the scaling process is not performed, but the data complement memory in the scaling block is accessed. The scaling process can guarantee up to 1/8 pixel accuracy by a three-dimensional convolution operation using a sampling function instead of simple thinning and straight line interpolation. The scaling process in the sub-scanning direction is performed by mechanical scaling. That is, enlargement / reduction is performed by changing the moving speed of the line scanner.

The processing so far is an image processing mainly due to the reading system. Inverse correction of the non-linear characteristics of the reading system is necessary.
The γ conversion by the γ conversion unit 6 which combines the non-linear characteristic conversion of the writing system and the notch characteristics of input and output is provided in the RAM by data download.

With respect to the read characteristics and the write characteristics, the binarized section 7 binarizes the optimized signal. Fixed binarization,
The multilevel signal is converted into a binary signal by, for example, tracking binarization using an average value in the block. The LD control unit 8 performs synchronization adjustment, power adjustment, lighting time adjustment, and the like of a writing system for using the binary signal as an LD lighting signal.
The LD 9 is turned on based on the signal under the LD control, and a latent image is formed on a photoconductor (not shown). Then, in an image forming unit (not shown), an electrophotographic image is formed on the paper by developing and fixing process.

With respect to these functional units, a density change that matches the amount of sensation intended by the user is performed in selecting a density notch on the operation unit 10. To change the density notch, the control unit 11 controls the MTF correction unit 4, the γ conversion unit 6, and the binarization unit 7 in conjunction with each other. In the conventional processing apparatus, the density change is performed by switching the γ curve.

FIG. 4 shows an example of document input / output characteristics (I / O characteristics) for generating γ data to be downloaded to the γ conversion section 6 in FIG. The horizontal axis of the graph is the density of the original to be read, and is the (original) input density to the image processing apparatus. The vertical axis indicates the formed electrophotographic density after the processing output, and is the (reproduced) output density from the image processing apparatus.

The I / O characteristic of (1) is a "standing γ characteristic".
Thus, the output density saturates faster than the input density. Although gradation reproducibility is poor, black and white can be clearly reproduced. On the other hand, the I / O characteristic of (2) is the “sleeping γ characteristic”, and the output density changes slowly with respect to the input density. The reproduction characteristic emphasizes the gradation.

In each graph, the characteristic of the density notch is reproduced by a curve of parallel movement. The numbers written on the parallel lines on the graph indicate density notches, 1 indicates density notch 1,
7 represents a density notch 7. In this mode, the notch 1 reproduces the read document with the highest density, and the notch 7 reproduces the read document with the lowest density. When thin characters such as pencils are clearly reproduced in black and white, the I / O characteristics of (1) are used for generating γ data. When the background stain is skipped, the I / O characteristic of (2) is used for generating γ data.

FIG. 6 shows an outline of generating γ data to be downloaded to the RAM. As described above, the conversion characteristics of the γ conversion unit 6 in FIG.
Download data that combines O characteristics (including notch characteristics). The first quadrant of the graph is an input / output characteristic indicating the read original density versus the output copy density, and has a density notch characteristic corresponding to FIG. In the second quadrant, after the LD control function,
9 shows the write characteristics of the output copy density for a digital signal including the write process.

The fourth quadrant shows an optical system for input document density and reading characteristics for digital signal values including shading correction. Lines connecting the intersections of the extension lines extending from the second and fourth quadrants on the third quadrant have γ conversion characteristics and are downloaded to the RAM. A γ conversion line segment is created in the third quadrant for each density notch in the first quadrant.

If the I / O characteristic is changed, the γ characteristic changes. In terms of parameters, when notch curves of the same I / O characteristic are moved in parallel at equal intervals, the density information of the image density also changes in parallel. The translation on the parameter is not always a translation in terms of the amount of human perception.

FIG. 7 shows some of the coefficients of the MTF correction filter of FIG. (1) shows a one-dimensional frequency characteristic. The frequency band to be emphasized is higher in f2 than in f1, and the sharpness is further emphasized. For example, a filter coefficient with a 5 × 5 matrix size shown in (2) has a higher degree of enhancement than the filter coefficient shown in (3), and enhances sharpness. In the MTF correction unit 4 in FIG. 1, an output can be arbitrarily selected from a plurality of correction filter coefficients.

FIG. 14 shows a circuit configuration of the filter portion.
In the example of 2 lines × 3 pixels, the pixel data is delayed by one clock by the delay element (D) 71. In the main scanning direction, image data up to one pixel delay and up to two pixels delay are generated. In the sub-scanning direction, data delayed by one line by the H delay line is generated. A matrix is generated from the delayed image data for both the main and sub-pixels, and a weighting unit 7 assigns a weight to each pixel between the corresponding positions of the matrix coefficients shown in FIG.
The sum is calculated in SUM73. The total amount becomes weighted correction data from peripheral pixels,
The sharpness of data corresponding to the contour portion of the image is enhanced.

FIG. 10 shows a parameter control mechanism linked to the density notch in FIG. N parallel two-dimensional filters 41 having various spatial frequency characteristics are processed. For example, for matrix data delayed for pixels, FIG.
The result of independently calculating the filter coefficient shown in (2) and the filter coefficient shown in (3) is obtained. According to the setting from the control unit 11, one of the processing results is selected by the selector 4.
Select with 2. This is the switching of the filter unit. In the γ-conversion unit 6, the density is changed by moving the density notch, but the download data is also switched by changing the type of the γ curve.

For example, according to the setting from the control unit 11, data having different γ gradients in FIG. 4 (1) or (2) is switched. Also in the binarization unit 7 of FIG. 1, for example, the threshold value of the fixed binarization is switched by the setting from the control unit 11.
These three switching signals are linked to the density notch. For example, in the character mode, the density notch
Reference numeral 4 designates a mode in which the copy output is made dark and a light density like pencil writing is clearly reproduced. Conversely, the notches 5 to 7 are assigned to a mode for reducing the copy output and removing the stain on the background of the read document.

In the notch mainly composed of a pencil, the γ characteristic to be used is “standing γ”, and FIG. 4A is downloaded to the RAM of the γ conversion unit 6 in FIG. The MTF filter uses the filter shown in FIG. 7 (2) which has strong emphasis characteristics in order to enhance the outline of a character portion and enhance sharpness. The binarization threshold is decreased from notch 4 to notch 1 so that the density becomes higher. As the density notch, notches 1 to 4 of the selected γ data are used.

As a result, the edge component is sufficiently emphasized by the strong MTF correction filter at the notch 1, and the solid portion is quickly saturated by the notch that rises first, and the binarization of the γ characteristic, which has a steep characteristic at the rising edge, reduces the binarization. A dark density reproduction image is formed by the threshold value.

Conversely, when the background portion is removed, the gamma characteristic is "sleepy gamma", and FIG. 4 (2) shows the RAM of the gamma conversion unit 6 in FIG.
To download. As the MTF filter, the filter shown in FIG. 7 (3) having a weak emphasis characteristic is used so as not to emphasize the noise component of the background stain portion. The binarization threshold is increased from the notch 5 to the notch 7 so that the density of the noise component in the background portion is deleted. As the density notch, notches 5 to 7 of the selected γ data are used.

Thus, the notch 7 does not emphasize the noise component of the background portion by the weak MTF correction, removes the low density portion by the γ characteristic, and sets the higher binarization threshold to the light background stain other than the word processing printed character. A reproduced image from which the image has been removed is formed.

FIG. 11 shows a mechanism for downloading data to the RAM. The RAM 51 has an address bus,
Data input bus, data output bus, chip select (C
S), write enable (WE), and output enable (OE) terminals. The γ data is downloaded from the data input terminal of the RAM 51 via the CPU 52 of the control unit 11, and the address terminal of the RAM 51 is used by switching the path between data download and normal γ conversion processing.

At the time of download, the address bus from the CPU 52 is selected, and at the time of normal use, image data for γ conversion is input. When the γ characteristic is switched and when the notch is switched, the corresponding converted data is reloaded into the RAM. Data can be changed programmably.

FIG. 2 is an overall block diagram of an image processing apparatus according to a second embodiment of the present invention. This image processing device reproduces the density in the binary photo mode for each density notch,
The filter coefficient, the γ table, and the binary dither pattern are linked to realize a variation that matches the user's feeling. That is, when a dark density notch is selected, a silver halide photographic original is clearly reproduced, and when a light density notch is selected, a moiré due to a halftone dot pitch of a print original is eliminated.

As in the first embodiment shown in FIG.
CD1, A / D converter 2, shading correction unit 3, main scanning scaling unit 5, γ conversion unit 6, LD control unit 8, LD9, operation unit 10, control unit 11, smoothing processing unit 21, dither processing A part 22 is provided.

Next, the operation will be described. A document on a document table (not shown) is irradiated with light from a lamp to read an image line by line. The read optical signal is input to the CCD 1 for each line, and converts the optical signal into an electric signal.
The output signal from the CCD 1 is converted in the A / D converter 2 into, for example, an 8-bit digital signal. Although the number of quantization steps is not limited to 8 bits, in the present embodiment, quantization is performed to 8 bits.

In order to correct illuminance unevenness due to the position of the optical reading system, shading correction is performed by the shading correction unit 3. The shading correction reads an absolute white reference white board and stores the data in a line memory. The original read image data of each line is normalized by the reference data stored in the line memory.

Regarding the data after the shading correction, A
In order to remove sampling distortion at the time of / D conversion and moire components due to halftone dot pitch in the document, high frequency components are removed by a smoothing filter of the smoothing processing unit 21. With respect to the image data that has been subjected to the smoothing processing, the main scanning conversion unit 5 performs electric scaling for enlargement / reduction.

In the case of the same magnification, the scaling process is not performed, but the data complement memory in the scaling block is accessed. The scaling process can guarantee up to 1/8 pixel accuracy by a three-dimensional convolution operation using a sampling function instead of simple thinning and straight line interpolation. The scaling process in the sub-scanning direction is performed by mechanical scaling. That is, enlargement / reduction is performed by changing the moving speed of the line scanner.

The processing up to this point is mainly image processing caused by the reading system. Inverse correction of the non-linear characteristics of the reading system is necessary.
The γ conversion by the γ conversion unit 6 which combines the non-linear characteristic conversion of the writing system and the notch characteristics of input and output is provided in the RAM by data download.

With respect to the reading characteristics and the writing characteristics, the binarizing section 7 binarizes the signal optimized. The dither processing unit 22 performs pseudo multi-level conversion on the multi-level signal by area gradation. For example, pseudo multi-value conversion using a binary signal by binary dither processing with an 8 × 8 matrix size is performed.

The LD control unit 8 performs synchronization adjustment, power adjustment, lighting time adjustment, and the like of a writing system for using the binary signal as an LD lighting signal. The LD 9 is turned on based on the signal under the LD control, and a latent image is formed on a photoconductor (not shown). Then, in an image forming unit (not shown), an electrophotographic image is formed on the paper by developing and fixing process.

With respect to these functional units, a density change that matches the amount of sensation intended by the user is performed in selecting a density notch on the operation unit 10. To change the density notch, the control unit 11 controls the smoothing processing unit 21, the γ conversion unit 6, and the dither processing unit 22 in conjunction with each other. In the conventional processing apparatus, the density change is performed by switching the γ curve.

FIG. 5 shows an example of document input / output characteristics (I / O characteristics) for generating γ data to be downloaded to the γ conversion section 6 in FIG. The horizontal axis of the graph is the density of the original to be read, and is the (original) input density to the image processing apparatus. The vertical axis indicates the formed electrophotographic density after the processing output, and is the (reproduced) output density from the image processing apparatus.

Both (1) and (2) are different from the character I / O characteristics in that they have an S-shaped characteristic in order to enhance the gradation reproducibility. It has a wide density reproduction range from low density to high density. The characteristic between the notches is not parallel movement as in the character mode, but the S-shaped curve shape is deformed so that the notch 1 and the notch 7 have a wide dynamic range.

(1) is for a silver halide photograph having a wide input density range.
The rise of γ is set to the low concentration part. Reproduce from the light density gradation of silver halide photographs. In (2), as compared with (1), the rise of the γ curve is shifted to the high density portion of the input document. The density corresponding to the background smear of the printed document is not reproduced. Since the printing ink does not have a density range and reproduces the density by area gradation, an input of a density lower than the density of the ink is regarded as a background stain.

The conversion data to be downloaded to the gamma conversion RAM of FIG. 2 is created based on the geometric conversion procedure of FIG. 6, similarly to the gamma data for characters. The difference from I / O
The I / O characteristics for photographs in FIG. 5 are used for the O characteristics instead of the I / O characteristics for characters in FIG. Data download to the RAM is also performed by switching the address bus in FIG. 11, as in FIG.

FIG. 8 shows a part of the coefficients of the smoothing filter of FIG. (1) shows a one-dimensional frequency characteristic,
It can be expressed by the characteristics of a low-pass filter, and fc1 and fc2
Is the cutoff frequency, the filter indicated by fc1 has higher smoothness than the filter indicated by fc2.
(2) shows an example of coefficients of a smoothing filter in a 5 × 5 matrix size. With this smoothing characteristic, high frequency components corresponding to moiré components are removed.

FIG. 9 shows an example of a dither matrix in the dither processing section of FIG. The properties of the formed image are changed by the dither pitch. By forming coarse halftone dots, it is possible to form an image that is strong against writing jitter, or to form fine halftone dots, thereby forming an image with high resolution by changing the dither pattern.

The dither pattern of the 8 × 8 matrix size shown in (1) generates a coarse halftone image corresponding to 70 lines. The dither pattern with the 8 × 8 matrix size shown in (2) forms a fine halftone dot pattern equivalent to 180 lines. The dither pattern is switched and used according to the characteristics of the input document and the reproduced image.

Interlocking switching of parameters in the photograph mode will be described with reference to FIG. The filter group is composed of filters 41 having a plurality of frequency characteristics, and performs respective parallel processing. For silver halide photographs, the cut-off frequency may be high because there is no halftone dot pitch on the document and sampling distortion during A / D conversion can be removed. Alternatively, an MTF correction filter that is not smooth but has a weak enhancement characteristic may be used.

In silver halide photography, a filtering process is performed so as not to deteriorate sharpness. A smoothing process is performed on the printed document in order to remove moire caused by the halftone dot pitch. The γ conversion section 6 selects from a plurality of data having the S-characteristic as shown in FIG. 5 and downloads the data to the RAM. Several kinds of γ data are prepared according to the dynamic range and the rising characteristics of the density conversion area. The dither processing unit 22 selects dither matrices having different numbers of lines. Like the γ data, the dither pattern is downloaded to the RAM and is switched programmably.

For example, the notch 1 to the notch 4 are used mainly for processing silver halide photographs, and the notches 5 to 7 are used mainly for printing originals. The cutoff frequency is increased from notch 7 toward notch 1. For the notch 1, a weak MTF correction filter is selected. At the notch 7, a filter having a low cutoff frequency and a strong smoothing characteristic is selected.

Regarding gamma, notch 1 to notch 4
In FIG. 5, a γ table corresponding to FIG. 5A having a fast rising characteristic is downloaded, and a density conversion curve from one notch to four notches is used. Notch 5 to Notch 7
In FIG. 5, the rising density was shifted for background removal.
A γ table corresponding to (2) is downloaded, and a density conversion curve from 5 to 7 notches is used. Further, the dither processing section 22 makes the dither pattern finer from the notch 7 toward the notch 1. The notch 1 reproduces the silver halide photograph sharply, and the notch 7 reproduces the printed original softly. Sharpness and softness are linked to density notch.

FIG. 3 is an overall block diagram of an image processing apparatus showing a third embodiment of the present invention. As in the second embodiment shown in FIGS. 1 and 2, the CCD 1, the A / D converter 2, the shading correction unit 3, the smoothing unit 21, the MTF
A gradation processing unit 31 and a phase control unit 32 are provided in addition to the correction unit 4, the main scanning scaling unit 5, the γ conversion unit 6, the LD control unit 8, the LD 9, the operation unit 10, and the control unit 11.

Next, the operation will be described. A document on a document table (not shown) is irradiated with light from a lamp to read an image line by line. The read optical signal is input to the CCD 1 for each line, and converts the optical signal into an electric signal.
The output signal from the CCD 1 is converted in the A / D converter 2 into, for example, an 8-bit digital signal. Although the number of quantization steps is not limited to 8 bits, in the present embodiment, quantization is performed to 8 bits.

To correct uneven illuminance due to the position of the optical reading system, shading correction is performed by the shading correction unit 3. The shading correction reads an absolute white reference white board and stores the data in a line memory. The original read image data of each line is normalized by the reference data stored in the line memory.

Regarding the data after shading correction, A
In order to remove sampling distortion at the time of / D conversion and moire components due to halftone dot pitch in the document, high frequency components are removed by a smoothing filter. The filter characteristic in this case is a smoothing process in which a band pass characteristic is provided and a spatial frequency of 133 lines to 200 lines corresponding to the halftone dot pitch of the printed document is removed.

High-frequency components of 200 lines or more are not removed.
With respect to the image data smoothed by the band-pass characteristic, MTF correction of a line image and a character portion is performed. Since the outline of the line drawing has a higher frequency component than the spatial frequency equivalent to 200 lines, it has not been removed by the previous smoothing processing. This high frequency component is emphasized by an MTF correction filter. With respect to the image signal that has been subjected to the filter processing, the main scanning conversion unit 5 performs electric scaling for enlargement / reduction.

In the case of the same magnification, the scaling processing is not performed, but the data complement memory in the scaling block is accessed. The scaling process can guarantee up to 1/8 pixel accuracy by a three-dimensional convolution operation using a sampling function instead of simple thinning and straight line interpolation. The scaling process in the sub-scanning direction is performed by mechanical scaling. That is, enlargement / reduction is performed by changing the moving speed of the line scanner.

The processing so far is an image processing mainly due to the reading system. Inverse correction of the non-linear characteristics of the reading system is necessary.
The γ conversion by the γ conversion unit 6 which combines the non-linear characteristic conversion of the writing system and the notch characteristics of input and output is provided in the RAM by data download.

With respect to the reading characteristics and the writing characteristics, the binarizing section 7 binarizes the signal optimized. An error diffusion process is performed on the multi-level signal in the tone processing unit 31 in order to maintain tone reproduction and sharpness of a character portion. With respect to the image data after the gradation processing, a phase signal in the main scanning direction is added in the phase control unit 32 and
It converts the power and phase of the M modulation signal and the PWM modulation signal.

The LD controller 8 performs synchronization adjustment, power adjustment, lighting time adjustment, and the like of a writing system for using the modulated signal as an LD lighting signal. The LD 9 is turned on based on the signal under the LD control, and a latent image is formed on a photoconductor (not shown). Then, in an image forming unit (not shown), an electrophotographic image is formed on the paper by developing and fixing process.

With respect to these functional units, a density change that matches the amount of sensation intended by the user is performed in the selection of the density notch on the operation unit 10. The control unit 11 controls the MTF correction unit 4, the γ conversion unit 6, the gradation processing unit 31, and the phase control unit 32 in conjunction with each other to change the density notch. In the conventional processing apparatus, the density change is performed by switching the γ curve.

FIG. 12 is a block diagram showing a mechanism of error diffusion processing as gradation processing. This block includes a memory 61, an error matrix 62, an error initialization unit 63, an error calculation unit 64, a selection unit 65, an error calculation unit 66, and a code conversion unit 67. Also, the phase control unit 32 is shown.

FIG. 13 is a diagram showing a pixel (2) continuous with the error matrix (1) in FIG. A quantization error at the time of requantization from surrounding pixels is distributed, and as a result of the distribution, the value obtained by summing the input original image data and the error from the surroundings is requantized. The requantization error is stored in the memory 61, and the calculation of the distribution value from the stored error signal is performed as shown in FIG.
Weight calculation is performed using a 3 × 6 error matrix shown in (1).
The values of the matrix coefficients are examples.

The position of * in the error matrix is the pixel currently being referred to. One line above the line is a coefficient group for the image one line before, and one line above is a coefficient group for the image one line before. . In order to reduce the influence of the surrounding errors in the sub-scanning direction, noisy pixels may be added at places where black changes to white like horizontal lines, and the sharpness of line segments is impaired.

For this reason, the error matrix size of 3 × 5 and the data distribution of the corresponding input original data are observed, and the current line is all white pixels, and the data before the first and second lines are all at a level that can be regarded as black. At this time, the total error information is cleared. In the state observation of the input original data, it is determined from the line memory image for filter processing that the line is the next line of the continuous black pixel area, and the determination information is used in the error diffusion processing unit.

The image data subjected to the error diffusion processing and requantized is code-converted so as to correspond to the dynamic range of the LD power modulation level. On the other hand, the phase control detects the leftmost pixel or the rightmost pixel of the continuous black pixel column from the density gradient of the continuous pixels in the main scanning direction, and in the case of the leftmost pixel, forms a dot in the right phase and generates a discontinuous pixel. Suppress. Similarly, in the case of the rightmost pixel, a dot is formed in the left phase.

The phase generation is shown in FIG. In pixels that are consecutive in the order of D0, D1, D2, and D3 from the left side, in the resolution-oriented 1-dot processing, D0 is set to the right phase,
3 is the left phase. Also, in the two-dot processing emphasizing gradation, an average value is obtained between two adjacent pixels, the density signal is replaced with an average value signal, and the phase signal is represented by the right phase of the averaged left pixel and the left phase of the right pixel. To process. D0 and D1, D2 and D
3, D0 and D2 are right phase, D1 and D3
Is a left phase pixel.

The continuous switching of parameters in the multi-value character / photo mode will be described with reference to FIG. The filter group is composed of filters 41 having a plurality of frequency characteristics, and performs respective parallel processing. In the smoothing processing unit 21, filters having different band-pass characteristics and the MTF correction unit 4 are used.
Then, filters having different emphasis frequency bands are prepared. The γ-conversion unit 6 selects from a plurality of data having the S-characteristics as shown in FIG.

Several types of γ data are prepared according to the dynamic range and rising characteristics of the density conversion area. In the gradation processing section 31, a plurality of error matrices having different weighting coefficients can be selected. The phase control unit 32 switches the phase generation in one-dot processing or the phase generation in two-dot processing.

The interlocking of the parameter switching with the density notch is, for example, that the notch 1 to the notch 4 are for processing mainly of a character portion in a mixed document of characters and pictures, and the notch 5 to the notch 7 are for the pattern section of the same document. As going from the notch 1 to the notch 7, the smoothing processing unit 21 changes the filter from the band pass characteristic to the low pass characteristic. In the MTF processing unit 4, the emphasis frequency band is expanded from the notch 7 toward the notch 1. For notch 1, a strong MTF correction filter is selected. At the notch 7, a filter having a weak emphasis characteristic is selected.

As for the gamma, at the notch 1 to the notch 4, the γ table corresponding to FIG. 5A having the fast rising characteristic is downloaded, and the density conversion curves from the first notch to the four notches are used. In the notch 5 to the notch 7, the rising density was shifted for background removal.
A γ table corresponding to (2) is downloaded, and a density conversion curve from 5 to 7 notches is used.

The error matrix 62 of the error diffusion processing section is
In the character-based density notch, the weight near the reference pixel is increased so as to expedite error generation. In a notch mainly composed of a picture, the weight coefficient is made flat in the matrix. The phase control switches to a one-dot processing phase for a character-based notch and a two-dot processing phase for a picture-based notch. Notch 1
, The sharpness of the character area is preferentially reproduced, and the notch 7 preferentially reproduces the smoothness of the picture portion. The sensory amount of sharpness and smoothness is linked to the density notch.

[0079]

According to the first aspect of the present invention, a reproduced image in the binary character mode can be obtained by setting a density variation that follows a user's sense amount from a density notch on the operation unit. In addition, a pencil document and a background stained document can be reproduced into a desired image only by a notch change in the same processing mode.

According to the second aspect of the present invention, a reproduced image in the binary photograph mode can be obtained by setting a density variation that follows a user's sense amount from a density notch on the operation unit. In addition, a silver halide photographic document and a printed document can be reproduced to a desired image only by a notch change in the same processing mode.

According to the third aspect of the present invention, a multi-valued character
A reproduced image in the photograph mode can be obtained by setting a density fluctuation that follows a user's sense amount from a density notch on the operation unit.

[Brief description of the drawings]

FIG. 1 is an overall block diagram of an image processing apparatus according to a first embodiment of the present invention.

FIG. 2 is an overall block diagram of an image processing apparatus according to a second embodiment of the present invention.

FIG. 3 is an overall block diagram of an image processing apparatus according to a third embodiment of the present invention.

FIG. 4 is a diagram showing an example of input / output characteristics for creating a character γ table.

FIG. 5 is a diagram showing an example of input / output characteristics for creating a picture γ table.

FIG. 6 is a diagram showing a conversion chart for creating a γ table.

FIG. 7 is a diagram illustrating an example of an MTF correction filter.

FIG. 8 is a diagram illustrating an example of a smoothing filter.

FIG. 9 is a diagram illustrating an example of a dither matrix.

FIG. 10 is a control block diagram for switching between a filter and a γ table.

FIG. 11 is a block diagram of data access to a RAM.

FIG. 12 is a block diagram illustrating a mechanism of error diffusion processing as gradation processing.

FIG. 13 is a diagram showing pixels continuous with the error matrix in FIG. 12;

FIG. 14 is a block diagram illustrating a circuit configuration of a filtering process.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 CCD 2 A / D converter 3 Shading correction part 4 MTF correction part 5 Main scanning magnification change part 6 γ conversion part 7 Binarization part 8 LD control part 9 LD 10 Operation part 11 Control part 21 Smoothing processing part 22 Dither processing Unit 31 gradation processing unit 32 phase control unit 41 two-dimensional filter 51 RAM 52 CPU 61 memory 62 error matrix 63 error initialization unit 64 error calculation unit 65 selection unit 66 error calculation unit 67 code conversion unit 71 delay element (D) 72 Weighting unit 73 SUM

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁶ Identification code FI H04N 1/403 H04N 1/40 103A

Claims (3)

[Claims] 1. An image reading means for optically reading an image and converting it to an electric signal, an A / D conversion means for converting an analog signal to a digital signal, a shading correction means for correcting density distortion of a read image, and an optical MTF correction means for correcting MTF degraded at the time of reading, main scanning magnification means for electrically enlarging / reducing in the main scanning direction,
Γ conversion means for performing density conversion in consideration of density characteristics of a reading system, process characteristics of a writing system, and variable characteristics of a density notch, a binarization means for converting a multi-level image signal into a binary signal, LD to control driving of laser diode
Control means, optical writing means for creating a latent image on the photoconductor, image creating means for creating a latent image on paper, operation means for inputting an operation mode of processing, and execution of operation contents An image processing apparatus comprising: a control unit that controls each unit and simultaneously changes an MTF correction characteristic, a γ characteristic, and a binarization characteristic in conjunction with a density notch on an operation unit. 2. An image reading means for optically reading an image and converting it into an electric signal, an A / D conversion means for converting an analog signal into a digital signal, a shading correction means for correcting a density distortion of a read image, and a reading method. Considering smoothing means for removing moiré of the image, main scanning magnification means for electrically enlarging / reducing in the main scanning direction, density characteristics of the reading system, process characteristics of the writing system, and variable characteristics of the density notch Converting means for performing density conversion, dither processing means for simulating the gradation of a multi-level image signal, and an LD for controlling driving of a writing laser diode
Control means, optical writing means for creating a latent image on the photoconductor, image creating means for creating a latent image on paper, operation means for inputting an operation mode of processing, and execution of operation contents An image processing apparatus comprising: a control unit for controlling each unit and simultaneously changing a smoothing filter characteristic, a γ characteristic, and a dither characteristic in conjunction with a density notch on an operation unit. 3. An image reading means for optically reading an image and converting it into an electric signal, an A / D conversion means for converting an analog signal into a digital signal, a shading correction means for correcting density distortion of a read image, and reading. Smoothing means for removing moiré of an image, MTF correction means for correcting MTF degraded during optical reading, main scanning magnification means for electrically enlarging / reducing in the main scanning direction, and density characteristics of a reading system Γ conversion means for performing density conversion in consideration of process characteristics of a writing system and variable characteristics of density notch, and gradation processing means for performing error diffusion processing of a multi-valued image signal,
Phase control means for generating a phase signal for dot formation at the time of writing, LD control means for controlling the drive of a writing laser diode, optical writing means for creating a latent image on a photoreceptor, Image forming means for forming an image on
An operation unit for inputting an operation mode of the processing, and each unit is controlled to execute the operation content, and a smoothing filter characteristic, M
An image processing apparatus comprising: control means for simultaneously changing a TF correction characteristic, a γ characteristic, a gradation characteristic, and a phase characteristic.