JP2945022B2 - Image processing device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus for outputting image data to a digital copying machine, a facsimile, a printer, a filing medium, and the like.

2. Description of the Related Art When an image on a document is read by an optical reading unit having a photoelectric conversion element such as a CCD image sensor, the resolution of a character image is required to be improved, and the gradation of a photographic image is emphasized. For this reason, when binarizing image data, the mode of the image processing method is changed depending on the type of image, such as a character image, a transfer image, and a mixed text / photo image.

Problems to be Solved by the Invention In the case of an image in which characters and photographs are mixed, the ratio between the characters and the photograph, the size of the photograph, the size of the halftone dot, the pitch, etc. change. Therefore, there is no most suitable image processing mode, and there is a problem that the image quality is inferior.

Means for Solving the Problems In an image processing apparatus that performs image processing on input image data and outputs the image data, a character image mode, a transfer image mode,
Image mode selecting means for selecting one of the character photograph image modes, binarizing means for selecting one of the dither threshold and fixed threshold to binarize input image data, and binarizing the image data Error dispersing means for dispersing an error at the time of
A plurality of settings combining one of the binarizing means and one of the necessity of the error dispersing means are made in advance for each of the character image mode, the photograph image mode, and the character photograph image mode, and one is selected by an operation. And a binarizing unit and an error dispersing unit that operate in accordance with the setting selected in the setting table for the mode selected by the image mode selecting unit. And

For images with mixed text and photos, if there are many characters, the resolution can be increased by binarization using a fixed threshold, and if there are many photos, the gradation can be improved by binarization using a dither threshold.
In addition, when binarizing, it is possible to select whether or not it is necessary to disperse the error of the image data.
Image processing with good conditions can be performed according to the type of image.

Embodiment An embodiment of the present invention will be described with reference to the drawings. FIG. 5 is a side view showing the inside of the document reading apparatus. In the figure, reference numeral 1 denotes a contact glass on which a document is placed, and reference numeral 2 denotes a document reading unit. The original reading unit 2 includes a carrier 5 having a light source 3 for illuminating the original and a mirror 4, and a carrier 8 having mirrors 6 and 7.
Are moved in the sub-scanning direction along the contact glass 1,
Mirrors 9,1 in which the original image is fixedly arranged with mirrors 4,6,7
The light is reflected at 0, and an image is formed on the CCD image sensor 12, which is a photoelectric conversion element, by the imaging lens 11.

FIG. 6 is a side view showing the inside of the laser pre-renter P. A photoconductor 14 is rotatably mounted at the center of a housing 13 of the laser printer P. Further, an exposure device 18 that scans the outer periphery of the photoconductor 14 with laser light emitted from the laser light emitting unit 15 via the imaging lens 16 and the mirror 17 is mounted on an upper portion of the housing 13. Further, around the photoreceptor 24, a charging charger 19 for charging the outer periphery of the photoreceptor 14, and an electrostatic latent formed by scanning the portion charged by the charging charger 19 with a laser beam by the exposure device 18 are used. A developing unit 20 for developing an image,
A transfer charger 21 for transferring the toner image developed by the developing unit 20 onto a sheet, a separation charger 22 and a separation claw 23 for separating the sheet from the photoconductor 14, and a cleaning unit 24 for wiping residual toner on the photoconductor 14 And are arranged. Further, the housing 13 is provided with a rotatable belt 25 for transporting the transferred paper and a fixing roller 26 for holding the paper and fixing the transferred image.

Next, the image processing circuit is shown in the block diagram of FIG. The CCD image sensor 12 of the image reading unit 2 is driven by a sensor driver 27, and the output side of the CCD image sensor 12 is connected via an amplifier 28, an A / D conversion unit 29, and a shading correction unit 30. It is connected to the filter processing unit 31. The filter processing unit 31 includes three types of spatial filters that convert frequency characteristics according to three types of reading modes such as a character image mode, a photo image mode, and a text / photo image mode.
32, 33, and 34 are connected in parallel to the shading correction unit 30, and the spatial filter 32 uses a high-pass filter so as to correspond to the character image mode and increase the resolving power. Ideally, the frequency characteristic of the spatial filter 32 should be an inverse function of the input system frequency representing the MTF of the input system including the image reading unit 2, but in practice it is necessary to consider noise components. A filter that performs MTF correction with a smaller correction amount than the function is employed. Next spatial filter 33
Employs a low-pass filter in correspondence with the photographic image mode to emphasize the gradation. This spatial filter
As the frequency characteristic of 33, a frequency characteristic that does not cause moire fringes in an output image with respect to a halftone original (133 to 175 lines / inch) without reducing the resolution as much as possible is selected. The other spatial filter 34 corresponds to a character / photograph image mode in which characters and photographs are mixed, and a filter having an intermediate characteristic between the spatial filters 32 and 33 is selected. For example, spatial filter 32
A high-pass filter with a smaller correction amount than that, a low-pass filter that passes data with a frequency higher than that of the spatial filter 33, or a low-pass filter that slightly suppresses noise components close to the Nyquist frequency. A filter is given as an example. Further, in order to correspond to the character / photograph image mode, the spatial filter 34 may be omitted and the data that has passed through the shading correction unit 30 may be directly passed.

Each spatial filter 32, 33,
The selector 34 is connected to a selector 35 serving as an image mode selection means, and an output side of the selector 35 is sequentially connected to a main scanning scaling unit 36 and a γ conversion unit 37. The selector 35 is a means for an operator to select a reading mode on an operation panel according to a reading mode, or a device in which the document reading apparatus 1 itself automatically determines a reading mode from document information, or a laser printer, a facsimile or a disk. And the like depending on the output destination of the read image data such as a file device.

The γ-conversion unit 37 converts image data so as to correspond to copy density adjustment. The γ-conversion unit 37 includes a binarization unit 38 for binarizing the image data and a value of 10 per pixel. And a PWM encoding means 39 for PWM-modulating the image data, and an error dispersing means 40 for dispersing an error in the image data are connected in parallel. Further, a selector, which is a binarization method selection unit for selecting the binarization method of the binarization unit 38 into one of a method using a fixed threshold 41 and a method using a dither threshold 42
43 are provided.

Next, details of the error dispersion means 40 are shown in the block diagram of FIG. As is clear from this figure, a plurality of latches 44, 45, 46 for inputting image data and adders 47, 48, 49 are alternately connected in series, and a memory 50 is connected to the output side of the adder 49. ing. The output side of an adder 52 connected to the memory 50 via a latch 51 is connected in parallel to the adders 47, 48, 49, and the output side of the adder 52 is connected to a comparator 54 via a limiter 53. An adder 55 and a selector 56 are connected in parallel, and the adder 55 is connected to the adder 55 via a selector 56 and a latch 57.
Connected to 49 inputs. The comparator 35 constitutes a part of a binarizing means 38 for comparing the set threshold value TH with the image data to binarize the image data.

In such a configuration, the image reading apparatus 1 reads an image of a document at a density of 16 pixels / mm in the main scanning and sub-scanning directions. The data output from the CCD image sensor 12 is amplified by the amplifier 28, and subsequently converted into digital data by the A / D converter 29 with 64 gradations per pixel. At this time, the number of gradations is 2 to the power of n and is handled as an n-bit binary signal. In this embodiment, 6-bit data is used.
Thereafter, the shading correction unit 30 corrects the illuminance unevenness of the light source 3 and the variation of the sensitivity of the CCD image sensor 12, and then passes through the filter processing unit 31. At this time, the signal system that has passed through the filter processing unit 31 is selected by the selector 35 according to the type of the image to be read. That is, it is determined whether the reading mode is the character image mode, the photograph image mode, or the character photograph image mode.

The determined image data is scaled by the main scanning scaling unit 36. The following magnification methods are listed.

Zooming in the main scanning direction by changing the reduction ratio of the optical system including the lens of the image reading device 1 and zooming in which the reading density in the sub-scanning direction is changed by changing the moving speed of the optical system.

Magnification in the main scanning direction in which image data at a sampling point at the time of zooming is obtained by interpolation using image data at the same magnification, and zooming in the sub-scanning direction as in the preceding paragraph.

Magnification by performing two-dimensional interpolation simultaneously with main and sub scanning using image data at the same magnification.

 The scaling in this embodiment is the scaling described above.

Subsequently, the scaled image data is converted by the γ conversion unit 37. This conversion corresponds to an operation of changing the density of a copy in a digital copying machine, a facsimile, or the like. In some cases, the density is adjusted by changing a threshold value for binarization. It converts itself according to the density selected by the operator or the device itself by the conversion table. The image data converted by the γ conversion unit 37 is binarized by a binarization unit 38 into black or white. The binarization method is performed by selecting the selector 43. The result of this selection is as shown in Table 1.

In Table 1, even if the reading mode is the same, depending on the selection number, the selection between the unnecessary error dispersion and the required error dispersion is made. The selection of the selection number is made by operating a dip switch (not shown) or an operation panel (not shown). If the scanning lines of the image are not finer than the fineness of the mesh of the image, so-called moiré fringes are generated and the image is deteriorated. In such a case, error dispersion is performed. FIG. 3 shows the scanning of the CCD image sensor 28 as the main scanning direction,
FIG. 4 is an explanatory diagram showing a state in which an error of image data is dispersed by setting a moving direction of the optical system as a sub-scanning direction. Here, a plurality of pixels are shown with one cell as one pixel, but an error Er at the time of binarization is shown.
Are allocated to four pixels adjacent in the direction of the arrow in the figure. There are various weights for distribution depending on each system, and there are some distribution methods such as distribution to all surrounding pixels, but in this embodiment, distribution is performed in quarters as indicated by arrows in FIG. . In addition, the pixel is binarized to “1” or “0” at a certain threshold, and when 64 tones (0 to 63) as in this embodiment, when the threshold is set to 32, 0 ~
31 is “0” and 32 to 63 are “1”. It decides which level of the 64 gradations the binary “0” and “1” are, and sets the values to th0 and th1, respectively. The meaning of the error Er of the image data changes depending on the setting of these th0 and th1. . That is, the error
Er indicates that the binarization result of the image data d is “0”.
Er = d-th1, Er = d-th1 in the case of "1". th0
= 0, th = 63 is a general idea, but different settings are made depending on the image reading device or the idea. In this embodiment, th0 = 0 and th = 47.

In FIG. 2, the data input to the latch 44 is image data. If the lower image data shown in FIG. 3 is input, this image data is latched by the latches 44, 45, 46 one after another by the transfer clock DCK. Each time the data is latched, the error Er generated from the previous line is dispersed, and this error
Er is added by the adders 47, 48, and 49. The image data to which the error Er has been added three times is stored in the memory 50. Since the error Er is positive or negative, this image data is signed data.

When the scanning of one line is completed and the next line is scanned, the image data stored in the memory 50 is sequentially called, and an error Er generated from the image data of one pixel before is added by the adder 52. At this point, the image data has a total of four errors Er.
Is added, and therefore, the number of gradations may be out of the range of 0 to 63. Therefore, the limiter 53 returns the gradation to the range of 0 to 63. That is, when image data <0, image data = 0 is set, and when image data> 63, image data =
Set to 63. Then, binarization is performed using the threshold value TH set by the comparator 54, and from the binarization result, Er = image data (binary “0”) or Er = image data−47 (binarization “ 1)) with the selector 56, and this time difference Er is 1/4
Are distributed to surrounding pixels. Here the selector 56
If Er = 0 is selected, error variance can be eliminated. That is, FIG. 2 is a block diagram of the error dispersing means 40, and a selector 56 for selecting execution of the error dispersing means 40 in the error dispersing means 40 is connected as distribution processing selecting means.

Next, multi-value data processing corresponding to multi-value writing will be described. In the past, laser printers could only reproduce black or white for one pixel, but recently a method of expressing a plurality of gradations for one pixel has been proposed. One of the methods is a PWM in which a pulse width of a pulse signal for driving a laser diode of a laser printer is controlled so that one pixel has gradation.
In the present embodiment, the writing on the printer side is performed by the PWM coding means 39. That is, the image data converted by the γ conversion unit 37 in FIG.
The path is sent to the PWM coding means 39 separately from the path to and is coded there. Coding does not mean generating the pulse itself for driving the laser diode, but converting it into numerical data corresponding to a desired pulse width. The laser printer P receiving this data generates a pulse for actually driving the laser diode.

Accordingly, the image data DA2 binarized by the binarization unit 38 and the image data DAM encoded by the PWM encoding unit 39 are sent to the laser printer P in parallel, but do not coexist. That is, when one of the image data is valid, the other is invalid, and the output of the signal line is fixed to a high impedance or any level of “1” or “0”. Thus, the printer can receive both binary data and multivalued data. For example, if a multi-level code is set to 3 bits, invalid data is fixed to “0”, “111” is all black, “000” is all white, and a code is set with halftone between them. As shown in FIG. 4, the binary data DA2 from the binarizing means 38 is divided into three ORs.
The binary data DA2 and the multi-value data DAM can also be synthesized by connecting the multi-value data DAM from the PWM coding means 39 to each of the OR gates 58 on a bit-by-bit basis. . Both the binarization processing and the multi-value processing can be performed with a simple configuration according to the present embodiment.

As described above, when the image data read by the image reading unit 2 is binarized and output, the signal system passing through the filter processing unit 31 can be selected by the selector 35 according to the type of the image. Method using a fixed threshold value by the selector 43 according to the type of
You can choose any of the valuation methods. Therefore, as shown in Table 1, when reading a text photograph image,
If the number of characters is large, the resolution can be increased by binarizing the image with a fixed threshold, and if the number of photographs is large, the gradation can be enhanced by the binarizing method with a dither threshold. Moreover, it is possible to select whether or not to disperse the error Er of the image data at the time of binarization by the selector 57. Therefore, it is possible to suppress the occurrence of moiré fringes and to provide an image processing method with good conditions according to the type of image. Can be adopted.

Advantageous Effects of Invention As described above, the present invention provides an image processing apparatus that performs image processing on input image data and outputs the image data, and selects one of a character image mode, a transfer image mode, and a character photograph image mode. Selecting means, binarizing means for selecting one of the dither threshold value and fixed threshold value and binarizing the input image data, and error dispersing means for dispersing an error in binarizing the image data; For each of the character image mode, the transfer image mode, and the character photograph image mode, a plurality of settings combining one of the binarizing means and one of the necessity of the error dispersing means are made in advance, and one is selected by operation. And a binarizing unit and an error dispersing unit that operate in accordance with the setting selected in the setting table for the mode selected by the image mode selecting unit. For images with mixed text and photos, the resolution can be increased by binarization with a fixed threshold if the number of characters is large, and the gradation can be enhanced by binarization with a dither threshold if the number of characters is large. In addition, when binarizing, it is possible to select whether or not it is necessary to disperse the error of the image data, so that the occurrence of moiré fringes can be suppressed, and image processing with good conditions can be performed according to the type of image. Has an effect.

[Brief description of the drawings]

1 shows an embodiment of the present invention. FIG. 1 is a block diagram showing an image processing circuit, FIG. 2 is a block diagram showing an electronic circuit of error dispersion means and dispersion processing selection means, and FIG. FIG. 4 is an explanatory diagram showing a state in which data errors are distributed to surrounding pixels, FIG. 4 is a circuit diagram showing output gates for binary data and multi-value data, FIG. 5 is a side view showing the inside of the image reading device, FIG. 6 is a side view showing the inside of the laser printer. 2 image reading unit 12 photoelectric conversion element 31 filter processing unit 32 to 34 spatial filter 35 image mode selection unit 38 binarization unit 40 error variance Means, 43
... Binarization method selection means, 56... Distributed processing selection means

Claims (1)

(57) [Claims] 1. An image processing apparatus for performing image processing on input image data and outputting the image data, comprising: an image mode selecting means for selecting one of a character image mode, a photograph image mode, and a character photograph image mode; Binarizing means for binarizing the input image data by selecting one of the image data and the fixed threshold value; error dispersing means for dispersing an error in binarizing the image data; A setting table in which a plurality of settings in which one of the binarizing means and one of the necessity of the error dispersing means are combined is set in advance for each of the image mode and the character / photograph image mode, and one is selected by an operation Wherein the binarizing means and the error dispersing means operate in accordance with the setting selected in the setting table for the mode selected by the image mode selecting means. Image processing apparatus.