JPH0846781A - Image forming device - Google Patents

Abstract

PURPOSE:To obtain the image processing unit adopting the image processing system in which blur or deformation caused at a character part or the like is improved in the case of reducing the image. CONSTITUTION:An image from an original is photoelectric-converted and converted into a digital image signal by an A/D converter section 27. An image area identification section 24 identifies the image signal as to whether the image is an image having character property or an image having gradation property based on a characteristic of a spatial frequency. In the case of the background image having the gradation property, an interpolation calculation section 18 multiplies an interpolation coefficient as a weight to be averaged and picture elements of the image signal are substantially thinned to reduce the image. In the case of the image having character property on the other hand, part of the picture elements of the image signal is multiplied with a fixed weight larger than the interpolation coefficient and averaged by the weight. Thus, the density of a border image between characters or the like and background or the like is reduced more than the picture element density averaged simply and then deformed or blurred image of characters or the like is not caused.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image forming apparatus, and more particularly to an image forming apparatus such as a digital copying machine capable of processing an image.

[0002]

2. Description of the Related Art Although a digital copying machine can perform various and highly flexible image processing by digital processing, it is said that the recording image quality is still far below analog. For example, since a digital copying machine can handle an image as a digital electric signal, it is possible to reduce a very small image, which is impossible with an analog copying machine, but the image is deteriorated due to the signal processing. There is a problem. On the other hand, various methods have been studied in which important information included in the original image is faithfully reproduced even at the time of reduction processing to record a high quality image. In general, there is a method called “projection method” as image processing for reducing an image.

Generally, when performing image reduction processing,
Basically, there are as many reduction destination images as the reduction ratio of the reduction source image. Therefore, as a reduction method for obtaining the reduction destination image, there is a simple “thinning” reduction method of thinning out the reduction source image. This method can be easily implemented, but has a major drawback that, if the thinned-out pixels happen to contain important image information, that information will be lost. The “projection method” was conceived for the purpose of improving this.

In this projection method, first, the pixel group projected on the reduction destination image at the time of reduction processing is multiplied by a weighting coefficient corresponding to the projection area (which is "length" in the case of one dimension). The interpolation calculation for calculating the weighted average is executed, and the result is used as the image data of the reduction destination pixel. Therefore, since the information of all the pixels of the reduced original image is saved by this interpolation calculation, there is no loss of information, and the reduced image of the gradation image such as the photograph portion is reproduced smoothly.

[0005]

As described above, the projection method used in the past has many advantages, but on the other hand, the data of the character portion and the thin line portion are also projected on average, so that the character portion and the boundary are not projected. A signal portion having a high spatial frequency such as an edge of a part is deteriorated. That is, the amplitude is attenuated. This appears as a decrease in so-called edge rounding. As a result, there is a problem that blurring or blurring occurs regardless of the image portion that requires resolution.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to solve the above-mentioned situation, and to provide an image processing apparatus which employs an image processing system capable of improving blurring and blurring which occur in a character portion when an image is reduced. To do.

[0007]

According to the present invention, scanning means for optically scanning an original, a lens system for guiding light rays from the original scanned by the scanning means, and a reduction ratio for image formation are provided. Means for setting, photoelectric conversion means for converting the light beam guided by the lens system into an image signal, and means for identifying whether the converted image signal has a character property or a gradation property, When the image signal has a gradation property, the image signal is multiplied by a first coefficient according to the reduction ratio,
When a reduced image signal is generated by averaging with the sum of the first coefficients and the image signal has gradation characteristics and character characteristics, the image signal has a first coefficient corresponding to the scaling factor and a first coefficient larger than this coefficient. Means for generating a reduced image signal by multiplying by a coefficient of 2 and averaging by the sum of the first and second coefficients, and an image for forming a reduced image on the image forming medium according to the generated reduced image signal An image forming apparatus is provided including: a forming unit.

According to the invention, the scanning means for optically scanning the original, the lens system for guiding the light beam from the original scanned by the scanning means, and the means for setting the reduction magnification for image formation A photoelectric conversion means for converting the light beam guided by the lens system into an image signal; a means for identifying whether the converted image signal has a character property or a gradation property; When the image signal has a gradation property, the fixed coefficient is selected, and when the image signal has a gradation property, the selecting unit that selects a plurality of interpolation coefficients according to the reduction ratio, and when the image signal has the gradation property, This image signal is multiplied by an interpolation coefficient according to the reduction ratio and averaged by the sum of the interpolation coefficients to generate a reduced image signal. When the image signal has gradation characteristics and character characteristics, the reduction ratio is added to the image signal. Interpolation coefficient according to and this coefficient Means for generating a reduced image signal by multiplying by a large fixed coefficient and averaging with interpolation and the sum of the fixed coefficients; and image forming means for forming an enlarged image on the image forming medium according to the generated reduced image signal. An image forming apparatus is provided, which comprises:

Furthermore, a scanning means for optically scanning the original, a lens system for guiding the light beam from the original scanned by the scanning means, a means for setting a reduction magnification for image formation, and a lens system. A photoelectric conversion means for converting the guided light ray into an image signal, a means for identifying whether the converted image signal has a character property or a gradation property, and a case where the image signal has a character property. , Selecting a fixed coefficient and selecting a plurality of interpolation coefficients according to the reduction ratio when the image signal has a gradation property, and reducing to this image signal when the image signal has a gradation property Multiply the interpolation coefficient according to the ratio and average with the sum of the interpolation coefficients to generate a reduced image signal. If the image signal has gradation characteristics and character characteristics, the interpolation coefficient according to the reduction rate is added to the image signal. And a fixed factor larger than this coefficient To generate a reduced image signal by averaging with the sum of the first and second coefficients, and when the image signal has character characteristics, the image signal is multiplied by a fixed coefficient larger than the interpolation coefficient Image forming means for averaging with a sum of the above to generate a reduced image signal, and an image forming means for forming an enlarged image on an image forming medium according to the generated reduced image signal. A device is provided.

[0010]

In the boundary portion of the image having the character property and the gradation property, the image signal is multiplied by the interpolation coefficient according to the reduction ratio and a fixed coefficient larger than this coefficient, and the interpolation and the fixed coefficient are averaged. And a reduced image signal is generated. Therefore, the density of the image signal is reduced as compared with the case where the averaging process is simply performed using the interpolation coefficient. Therefore, blurring, blurring, and the like are prevented at the image-boundary portion between the character and the background, and the character or the like is reproduced clearly.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a control block diagram of the digital copying machine, FIG. 2 is a block diagram showing a configuration of a main part of the digital copying machine, and FIG. 3 is a sectional view showing an internal structure of the digital copying machine.

First, the structure of the digital copying machine will be described with reference to FIG. As shown in FIG. 3, the digital copying machine includes a scanner section c for optically reading image information of a document, and a printer for outputting an image read by the scanner section c onto a recording material, that is, a copy sheet. It has an engine d.

In the scanner section c, the original to be copied is placed on the original placing table 5, and the original p placed on the placing table 5 is a fluorescent lamp as a light source extending in the sub-scanning direction. The reflected light from the document p illuminated by the fluorescent lamp 29 is photoelectrically converted by the CCD sensor 26, and the reflected light is image information converted into image signals. A reflector for efficiently focusing the illumination light from the fluorescent lamp 29 on the original 6 is arranged on the side of the fluorescent lamp 29, and between the fluorescent lamp 1 and the CCD sensor 26, A light beam traveling from the document p to the CCD sensor 26, that is, a plurality of mirrors for bending an optical path through which a light beam reflected from the document passes, and for converging the reflected light on a light collecting surface of the CCD sensor 26. Lens etc. are arranged.

The exposure lamp 29 and an optical system for guiding the reflected light from the document to the CCD sensor 26 are mounted on a carriage and conveyed in the main scanning direction by a pulse motor (not shown).
The original area in the sub-scanning direction is illuminated by the fluorescent lamp 29, the areas in the sub-scanning direction of the original are illuminated one after another by moving the carriage in the main scanning direction, and the entire area of the original is illuminated by the fluorescent lamp 29. It will be.

A document is placed on the document table 5 above the document table 5.
A document cover 91a that is brought into close contact with is placed. The original document cover 91a is provided with a coordinate input section for designating a closed area on the original document p placed on the table 5 for image processing. The document pressing can be replaced with, for example, an SDF, a semi-automatic document feeder, or an ADF, an automatic document feeder, depending on the size of the digital copying machine or the copying capability.

The image forming section d is provided with a photosensitive drum 93 which is cylindrical and is rotated in a desired direction via a motor (not shown) and charged to a desired potential. When the photoconductor drum 93 is irradiated with the light beam, the potential of the region irradiated with the light beam is changed, and an electrostatic latent image is formed on the photoconductor drum 93.

Around the photosensitive drum 93, a charging device 94 for applying a desired potential to the drum 93, a laser unit 95 for outputting to the photosensitive drum 94 a laser beam modulated according to image information, and this laser unit. A developing device 96 that supplies a visualizing agent, that is, toner to the electrostatic latent image formed on the photoconductor drum 93 by the laser beam from the developing device 96 to develop it, and visualizes the photoconductor drum 94 developed by the developing device 96. A transfer device 97 that transfers the formed toner image to a recording material fed from a recording material feeding portion described later, that is, a copy sheet is arranged.

On the downstream side of the transfer device 97 in the rotation direction of the photosensitive drum 94, the toner remaining on the surface of the photosensitive member 93 of the photosensitive drum 94 is removed and the laser beam is applied to the photosensitive member 93. A cleaner unit 98 for erasing the generated potential change for the next image formation is arranged.

Between the developing device 96 and the transfer device 97, a recording material feeding portion 99 for feeding a copy sheet onto which the toner image formed on the photoconductor 93 is transferred toward the transfer device 97. Are arranged. Further, a fixing device 100 for fixing the toner image on the copy paper is provided in the direction in which the copy paper on which the toner image is transferred by the transfer device 97 is separated from the photoconductor 93. Between the fixing device 100 and the transfer device 97, a conveying device 10 for conveying a copy sheet toward the fixing device 100.
1 is arranged.

In such a digital copying machine, an image is processed by the control system shown in FIG. 1 and control of the image processing is executed. In FIG. 1, reference numeral 11 denotes a main control unit that controls the present digital copying machine. This main control unit 1
1 includes four sub-control units 12 to 1 that execute control of each unit.
4 is connected, and the main control unit 11 includes an operation panel 16,
It is connected to the area management unit 17 that manages the image processing area, the image quality improvement unit 22 that improves the image quality of the captured image, the image editing unit 19 that edits the image, and the image processing unit 20 that processes the image, and controls these. ing. Further, the main control unit 11 is provided with a buffer 21 for temporarily storing image information and a parameter setting unit 23 for setting parameters for interpolation calculation, and interpolation calculation for performing interpolation calculation of image information at the time of enlargement. The part 18 is connected. The parameter setting unit 23
The image improvement parameters set on the operation panel 16 are output to the interpolation calculation unit 18.

The sub-control section 12 controls the light source intensity of the fluorescent lamp 29, a mechanism driving section 25 that controls mechanical mechanisms such as the sheet feeding mechanism shown in FIG. 3, and detects reflected light rays. And a CCD sensor for converting the converted image signal into an image signal, that is, a photoelectric conversion unit 26, an A / D conversion unit 27 for converting the converted analog image signal into a digital signal, and a correction unit 28 for correcting the image signal such as shading. And are controlling these. Further, the light source control unit 24 is connected to an exposure lamp 29 which is a light source, the light intensity of the light source is controlled, and the mechanism driving unit 25 is connected to an input unit mechanism 30 such as a pulse motor for moving the carriage. Is driven. Therefore, the entire area of the document is illuminated with the appropriate illumination light.

The sub-control unit 13 expands the edited or processed image signal for image formation, and stores the image, and outputs the image from the image expansion unit 31 and the image expansion unit 31 as a laser modulation signal. An output unit 32, a detection unit 33 that detects the laser beam and adjusts the light intensity of the laser output from the semiconductor laser of the laser unit 95, and a mechanism drive unit 35 such as a drive system such as a motor and solenoid in the laser unit 95. It is connected to the driving mechanism drive unit 34 and controls them.

The sub-control unit 14 is connected to a coordinate input unit 36 for designating an image area to be captured and controls them. The sub control unit 37 is connected to the data transmitting / receiving unit and controls data transmission / reception with an external device. Area management unit 1
7 detects a closed area based on the coordinate information on the original image input from the coordinate input unit 36, and when the image signal is within the closed area, 7 indicates that each image signal is within the area. The signal shown is given. On the other hand, when the image signal is outside the closed region, a signal indicating that it is outside the region is given to each image signal.

In the above-mentioned digital copying machine, the original p is illuminated by the fluorescent lamp 29, and the reflected light beam reflected from the original is imaged on the CCD sensor 26 and converted into an analog electric signal. This analog image signal is A /
The digital signal is converted into a digital signal by the D conversion unit 27 and is supplied to the correction unit 28 for shading correction. This correction unit 2
In 8, the shading-corrected image signal is temporarily stored in the buffer 21 and output to the image improving unit 22. In the image improvement unit 22, the digital image signal is calculated by the digital filter, the characteristics of the image signal in the low frequency region and the characteristics of the high frequency region are improved, and the result is supplied to the interpolation calculation unit 18.

By operating the coordinate input section 36, an area on the document can be divided and designated, and information on the divided designated area is stored in the area management section 17. In the buffer 21, a character for identifying whether the area on the document designated by the coordinate input unit 36 is character information having a large density difference or clear darkness or background information including halftone such as a photograph is identified. / Image area identification unit 2 for generating a background identification signal
4 are connected. This identification signal is sent to the area management unit 17
Is sent to the image improving section 22 together with the area management information, and the image quality of the image signal is improved according to the identification signal. That is,
Character information having a high spatial frequency is enhanced in high-frequency characteristics and improved to sharp image information. In addition, the background information having a low spatial frequency has the low frequency characteristic emphasized, and is improved to image information having a gentle graduation.

The image area identifying section 24 analyzes the spatial frequency of the area on the document designated by the coordinate input section 36 and determines whether the information is character information having a high spatial frequency or background information having a low spatial frequency. Is determined. Specifically, as shown in FIG. 4, the image signal supplied from the buffer 21 is supplied to the digital filter 38 for extracting its high frequency component, and the high frequency component is supplied to the binarization circuit 39. The binarization circuit 39 compares the reference signal from the reference signal generator 40, and the comparison result is stored in the binarized image table 41. The stored binarized image is compared and referred to by the reference circuit 42 with a plurality of standard patterns, which are stored in advance in the pattern table 43 and include background patterns such as character patterns and photographs. The result of comparison and reference is the judgment circuit 44.
The determination result is given to the identification signal "1" or "
The image area discrimination circuit 21 generates 0 ".

The image information whose image has been improved is subjected to interpolation calculation in the interpolation calculation section 18 according to the identification signal and converted into a reduced image signal. That is, when the image to be reduced is character information with a high spatial frequency, the average calculation is performed and the thinned image signal is output to the image editing unit 19 as image information. On the other hand, when the image to be reduced is background information having a low spatial frequency, the image signal is interpolated according to the interpolation coefficient determined by the interpolation coefficient setting unit 23, and the image signal is thinned out. The enlarged image signal is output to the image editing unit 19 as image information. The interpolation calculator 18 will be described in more detail later with reference to FIG.

In the image editing section 20, the image information is edited according to the image editing information such as trimming and masking input to the operation panel 16. Further, the edited image information is subjected to various processing such as italicizing the character information in the image processing unit 20 as necessary, and is expanded into pixel information of one frame by the image expansion unit 31. The developed image is output to the image output unit 32 and printed as a visible image. That is, the laser beam output from the laser unit 95 is modulated in accordance with the developed image information to form a latent image on the photosensitive drum 94, which is visualized by the developing device 96 and transferred to the recording material. Is fixed on the recording material and output.

FIG. 2 shows a detailed block diagram of the interpolation calculation section 18. As shown in FIG. 2, the image signal supplied from the image improving unit 22, that is, the density data, is supplied to the register circuits 45 of the first to fourth shift registers 45A to 45D, and the identification signal is supplied from the first to fourth. Fourth
Of the flip-flops 46A to 46D. The image signal is, for example, a pixel signal including 8-bit density information. When this image signal is sequentially supplied to the register circuits 45 of the first to fourth shift registers 45A to 45D, the image signal is The shift order is from the fourth shift register 45D to the first shift register 45A. Similarly, the identification signal is also supplied to the circuits 46 of the first to fourth flip-flops 46A to 46D in synchronization with the image signal, and the fourth flip-flop 46D to the first flip-flop 46A are input according to the input order. Will be shifted towards. The first, second, third and fourth image signals are the first to fourth shift registers 45A.
To 45D, the fifth image signal is stored in the register circuit 4
5 is input to the first multiplier 47,
A, the second image signal is supplied to the second multiplier 47B, the third image signal is supplied to the third multiplier 47C, and the fourth image signal is supplied to the fourth multiplier 47D. The fifth image signal is supplied to the fifth multiplier 47E. Similarly, when the first, second, third, fourth, and fifth identification signals are supplied to the interpolation coefficient switching circuit 48 and the interpolation coefficient switching signal is supplied from the CPU 11, the corresponding first, second, and fifth signals are supplied. 3, 4th and 5th switching contacts 48A, 48B, 4
8C, 48D, and 48E are switched according to the identification signal. For example, if the first, second, and third identification signals are background identification signals “0” indicating the background of a photograph or the like, the first,
The second and third multipliers 47A, 47B, 47C are connected to the reduction interpolation coefficient calculation unit 23 via the switching contacts 48A, 48B, 48C, and the interpolation coefficients corresponding to the reduction ratio are the first, second, and The third and fourth multipliers 47A, 47B, and 47C are supplied to the third and fourth multipliers 47A, 47B, and 47C, and when the fourth and fifth identification signals are the character identification signal "0" indicating the background of characters and the like, the fourth and fifth multiplications. The devices 47D and 47E have switching contacts 48D and 48E.
Is connected to fixed values 50D and 50E via E, and a fixed fixed interpolation coefficient is applied to the fourth and fifth multipliers 47D and 47D.
Supplied to 7E. Further, the interpolation coefficient and the fixed value are supplied to the division coefficient calculation unit 49 and added.

In the first, second, third, fourth and fifth multipliers 47A, 47B and 47C, the image signal is multiplied by the correction coefficient and the fixed value, and the multiplication result is added by the adder 51. Further, in the graduation device 52, the addition output is gradually divided by the addition value of the correction coefficient and the fixed value. The result is stored in the memory 53 as a first reduced image. At this time, the memory 53 is stored in the memory area specified by the write address generated by the address generator 54. The address generator 54 generates an address at a timing according to the reduction ratio according to a command from the CPU 11. That is,
Even when the sixth image signal is input to the register circuit 45, the above-described calculation is performed on the second, third, fourth, fifth and sixth image signals, and the calculation result is the graduation unit. Although it is output from 52, the write address is not generated at this time. Therefore, the calculation result is not stored in the memory 53. In the case where the reduction ratio is 1/400, the third, fourth, fifth, sixth and seventh and fourth images are similarly inputted when the seventh and eighth image signals are inputted to the register circuit 45. , The fifth, sixth, seventh, and eighth image signals are subjected to the above-described calculation, and the calculation result is output from the gradual divider 52.
At this time, no write address is generated.

When the ninth image signal is input to the register circuit 45, the above-mentioned operations are performed on the fifth, sixth, seventh, eighth and ninth image signals, and the operation result is obtained. It is output from the graduation unit 52, and at this time, the write address is generated again. That is, when the fifth and ninth image signals are input, the write signal is generated and the reduced image signal is written in the memory 53. This substantially corresponds to the thinning of the 3-pixel signal from the 4-pixel signal.

The process in FIG. 2 will be described more specifically with reference to FIGS. 5 and 6. FIG. 5 shows a case where a background identification signal "0" representing a background is continuously generated as an identification signal. Further, FIG. 5 shows that when the reduction ratio is 1/400, the interpolation coefficients are 0.8, 1.0, 1.0, 1..x for the first, second, third, fourth and fifth pixels. An example of 0, 0.2 is shown. The interpolation coefficient is 0.
8, 1.0, 1.0, 1.0, 0.2 are multiplied, all the pixel densities multiplied by the interpolation coefficient are added, and the interpolation coefficients 0.8, 1.0, 1.0, 1 are added again. The density of one reduced pixel is determined by dividing by 4.0, which is a value obtained by adding 0.0 and 0.2. Similarly, when the background continues, similar processing is performed, so that the pixel density is gently changed and a smooth image is reproduced. On the other hand, FIG. 6 shows a case in which a background identification signal "0" representing a background and a character identification signal "1" representing a character or a thin line are generated as identification signals. Further, FIG. 6 similarly shows the first, second and third corresponding to the background identification signal "0" when the reduction ratio is 1/400.
Interpolation coefficients are set to 0.8, 1.0, 1.0 corresponding to the pixels, and fixed values for the fourth and fifth pixels are 1.0 ,. The example set to 10 is shown. The interpolation coefficient and the fixed value are respectively multiplied by 0.8, 1.0, 1.0, 1.0, and 1.0 with respect to the density of the pixel, and all the pixel density multiplied by the interpolation coefficient are added, Again the interpolation coefficient and fixed value 0.8, 1.0, 1.0,
The density of one reduced pixel is determined by dividing by 1.0, which is a value obtained by adding 1.0 and 4.2. By this processing, the preparation for processing the pixel such as the character having the next highest density is completed. That is, when the pixels such as characters are included, the density is usually higher than when the background density is calculated, but since the value to be divided is large, the reduced image has the same density. The density is kept low, and the density of the boundary portion between the background and the characters becomes low, and the following characters and the like are prevented from being blurred or crushed. Of course, following the pixels in FIG.
Pixels having a high density with an identification number of "1" are consecutive, but in this case, fixed values are 1.0, 1.0, 1.0, 1.0 for the density of the pixels, respectively. , 1.0 is added, all pixel densities multiplied by the interpolation coefficient are added, and the fixed value 1.0, 1.0, 1.0, 1.0, 1.0 is added again to obtain a value 5.0. As described above, the density of one reduced pixel such as a character is determined by dividing the image portion corresponding to the background portion by the background method in the present invention. A reduced image without is reproduced, and in the image part with high spatial frequency such as the edge of the character part and the boundary part,
Image data is not uniformly averaged but processed by setting a large weight, so the pixel density is reduced, characters with high pixel density become clear, and the edges of character parts and boundaries are blunted. Is prevented. Therefore, according to the present invention, it is possible to improve the blurring and the blurring that occur in the character portion or the like during the reduction.

[0033]

According to the present invention, it is possible to obtain a magnified image in which blurring does not occur even at the edges of the character portion and the boundary portion while maintaining the smooth image reproduction in the gradation image portion.

[Brief description of drawings]

FIG. 1 is a block diagram showing a control system of an image forming apparatus according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a circuit configuration of an interpolation calculation unit shown in FIG.

3 is a sectional view schematically showing the structure of the image forming apparatus shown in FIG.

FIG. 4 is a block diagram showing a circuit example of an image area identification section shown in FIG.

FIG. 5 is an explanatory diagram schematically showing a method of performing a projection process on an image signal for reducing a background image.

FIG. 6 is an explanatory diagram schematically showing a method of increasing a weight given to an image signal in order to reduce an image of a boundary portion of a background and a character and performing projection processing.

[Explanation of symbols]

 5 ... Mounting table 11 ... Main control device 17 ... Area management unit 18 ... Interpolation calculation unit 19 ... Image editing unit 20 ... Image processing unit 21 ... Buffer 22 ... Image quality improvement unit 26 ... CCD sensor 24 ... Image area identification unit 28 ... Correction unit 31 ... Image development unit 36 ... Coordinate input unit 45 ... Register circuit 48 ... Switching circuit 49・ ・ ・ Division coefficient calculation unit 53 ・ ・ ・ Memory 47A to 47E ・ ・ ・ Multiplication circuit 51 ・ ・ ・ Addition circuit 52 ・ ・ ・ Division circuit 93 ・ ・ ・ Photosensitive drum 95 ・ ・ ・ Laser unit 96 ・ ・Development device 97 ... Transfer device 100 ... Fixing device

Claims (3)

[Claims] 1. A scanning unit for optically scanning a document, a lens system for guiding a light beam from the document scanned by the scanning unit, a unit for setting a reduction ratio for image formation, and a lens system for guiding. A photoelectric conversion means for converting the reflected light beam into an image signal, a means for identifying whether the converted image signal has a character property or a gradation property, and a case where the image signal has a gradation property. If the image signal is multiplied by a first coefficient according to the reduction ratio and averaged by the sum of the first coefficients to generate a reduced image signal, and the image signal has gradation characteristics and character characteristics, Means for multiplying the signal by a first coefficient and a second coefficient larger than this coefficient according to the scaling factor, and averaging by the sum of the first and second coefficients to generate a reduced image signal; A reduced image is formed on the image forming medium according to the reduced image signal. An image forming apparatus characterized by comprising an image forming means. 2. A scanning means for optically scanning an original, a lens system for guiding a light beam from the original scanned by the scanning means, a means for setting a reduction magnification for image formation, and a lens system for guiding. Photoelectric conversion means for converting the reflected light beam into an image signal, means for identifying whether the converted image signal has a character property or a gradation property, and when the image signal has a character property, When a fixed coefficient is selected and the image signal has a gradation property, a selection unit that selects a plurality of interpolation coefficients according to the reduction ratio, and when the image signal has a gradation property, the reduction ratio is added to this image signal. Is multiplied by an interpolation coefficient and averaged by the sum of the interpolation coefficients to generate a reduced image signal. When the image signal has gradation characteristics and character characteristics, the image signal has an interpolation coefficient corresponding to the reduction rate and Multiply a fixed coefficient larger than this coefficient Interpolation and averaging with a sum of fixed coefficients to generate a reduced image signal, and image forming means for forming an enlarged image on the image forming medium according to the generated reduced image signal. A characteristic image forming apparatus. 3. A scanning unit for optically scanning an original, a lens system for guiding a light beam from the original scanned by the scanning unit, a unit for setting a reduction ratio for image formation, and a lens system for guiding. Photoelectric conversion means for converting the reflected light beam into an image signal, means for identifying whether the converted image signal has a character property or a gradation property, and when the image signal has a character property, When a fixed coefficient is selected and the image signal has a gradation property, a selection unit that selects a plurality of interpolation coefficients according to the reduction ratio, and when the image signal has a gradation property, the reduction ratio is added to this image signal. Is multiplied by an interpolation coefficient and averaged by the sum of the interpolation coefficients to generate a reduced image signal. When the image signal has gradation characteristics and character characteristics, the image signal has an interpolation coefficient corresponding to the reduction rate and Multiply a fixed coefficient larger than this coefficient Interpolation and averaging with the sum of fixed coefficients to generate a reduced image signal, and if the image signal has character characteristics, multiply the image signal by a fixed coefficient larger than the interpolation coefficient and average with the sum of fixed coefficients. An image forming apparatus comprising: a unit that generates a reduced image signal; and an image forming unit that forms an enlarged image on an image forming medium according to the generated reduced image signal.